CaptureResult


public class CaptureResult
extends CameraMetadata<Key<?>>

java.lang.Object
   ↳ android.hardware.camera2.CameraMetadata<android.hardware.camera2.CaptureResult.Key<?>>
     ↳ android.hardware.camera2.CaptureResult


The subset of the results of a single image capture from the image sensor.

Contains a subset of the final configuration for the capture hardware (sensor, lens, flash), the processing pipeline, the control algorithms, and the output buffers.

CaptureResults are produced by a CameraDevice after processing a CaptureRequest. All properties listed for capture requests can also be queried on the capture result, to determine the final values used for capture. The result also includes additional metadata about the state of the camera device during the capture.

Not all properties returned by CameraCharacteristics#getAvailableCaptureResultKeys() are necessarily available. Some results are partial and will not have every key set. Only total results are guaranteed to have every key available that was enabled by the request.

CaptureResult objects are immutable.

Summary

Nested classes

class CaptureResult.Key<T>

A Key is used to do capture result field lookups with CaptureResult#get

Inherited constants

Fields

public static final Key<Boolean> BLACK_LEVEL_LOCK

Whether black-level compensation is locked to its current values, or is free to vary.

public static final Key<Integer> COLOR_CORRECTION_ABERRATION_MODE

Mode of operation for the chromatic aberration correction algorithm.

public static final Key<RggbChannelVector> COLOR_CORRECTION_GAINS

Gains applying to Bayer raw color channels for white-balance.

public static final Key<Integer> COLOR_CORRECTION_MODE

The mode control selects how the image data is converted from the sensor's native color into linear sRGB color.

public static final Key<ColorSpaceTransform> COLOR_CORRECTION_TRANSFORM

A color transform matrix to use to transform from sensor RGB color space to output linear sRGB color space.

public static final Key<Integer> CONTROL_AE_ANTIBANDING_MODE

The desired setting for the camera device's auto-exposure algorithm's antibanding compensation.

public static final Key<Integer> CONTROL_AE_EXPOSURE_COMPENSATION

Adjustment to auto-exposure (AE) target image brightness.

public static final Key<Boolean> CONTROL_AE_LOCK

Whether auto-exposure (AE) is currently locked to its latest calculated values.

public static final Key<Integer> CONTROL_AE_MODE

The desired mode for the camera device's auto-exposure routine.

public static final Key<Integer> CONTROL_AE_PRECAPTURE_TRIGGER

Whether the camera device will trigger a precapture metering sequence when it processes this request.

public static final Key<MeteringRectangle[]> CONTROL_AE_REGIONS

List of metering areas to use for auto-exposure adjustment.

public static final Key<Integer> CONTROL_AE_STATE

Current state of the auto-exposure (AE) algorithm.

public static final Key<Range<Integer>> CONTROL_AE_TARGET_FPS_RANGE

Range over which the auto-exposure routine can adjust the capture frame rate to maintain good exposure.

public static final Key<Integer> CONTROL_AF_MODE

Whether auto-focus (AF) is currently enabled, and what mode it is set to.

public static final Key<MeteringRectangle[]> CONTROL_AF_REGIONS

List of metering areas to use for auto-focus.

public static final Key<Integer> CONTROL_AF_SCENE_CHANGE

Whether a significant scene change is detected within the currently-set AF region(s).

public static final Key<Integer> CONTROL_AF_STATE

Current state of auto-focus (AF) algorithm.

public static final Key<Integer> CONTROL_AF_TRIGGER

Whether the camera device will trigger autofocus for this request.

public static final Key<Integer> CONTROL_AUTOFRAMING

Automatic crop, pan and zoom to keep objects in the center of the frame.

public static final Key<Integer> CONTROL_AUTOFRAMING_STATE

Current state of auto-framing.

public static final Key<Boolean> CONTROL_AWB_LOCK

Whether auto-white balance (AWB) is currently locked to its latest calculated values.

public static final Key<Integer> CONTROL_AWB_MODE

Whether auto-white balance (AWB) is currently setting the color transform fields, and what its illumination target is.

public static final Key<MeteringRectangle[]> CONTROL_AWB_REGIONS

List of metering areas to use for auto-white-balance illuminant estimation.

public static final Key<Integer> CONTROL_AWB_STATE

Current state of auto-white balance (AWB) algorithm.

public static final Key<Integer> CONTROL_CAPTURE_INTENT

Information to the camera device 3A (auto-exposure, auto-focus, auto-white balance) routines about the purpose of this capture, to help the camera device to decide optimal 3A strategy.

public static final Key<Integer> CONTROL_EFFECT_MODE

A special color effect to apply.

public static final Key<Boolean> CONTROL_ENABLE_ZSL

Allow camera device to enable zero-shutter-lag mode for requests with android.control.captureIntent == STILL_CAPTURE.

public static final Key<Integer> CONTROL_EXTENDED_SCENE_MODE

Whether extended scene mode is enabled for a particular capture request.

public static final Key<Integer> CONTROL_LOW_LIGHT_BOOST_STATE

Current state of the low light boost AE mode.

public static final Key<Integer> CONTROL_MODE

Overall mode of 3A (auto-exposure, auto-white-balance, auto-focus) control routines.

public static final Key<Integer> CONTROL_POST_RAW_SENSITIVITY_BOOST

The amount of additional sensitivity boost applied to output images after RAW sensor data is captured.

public static final Key<Integer> CONTROL_SCENE_MODE

Control for which scene mode is currently active.

public static final Key<Integer> CONTROL_SETTINGS_OVERRIDE

The desired CaptureRequest settings override with which certain keys are applied earlier so that they can take effect sooner.

public static final Key<Integer> CONTROL_VIDEO_STABILIZATION_MODE

Whether video stabilization is active.

public static final Key<Float> CONTROL_ZOOM_RATIO

The desired zoom ratio

Instead of using android.scaler.cropRegion for zoom, the application can now choose to use this tag to specify the desired zoom level.

public static final Key<Integer> DISTORTION_CORRECTION_MODE

Mode of operation for the lens distortion correction block.

public static final Key<Integer> EDGE_MODE

Operation mode for edge enhancement.

public static final Key<Integer> EXTENSION_CURRENT_TYPE

Contains the extension type of the currently active extension

The capture result will only be supported and included by camera extension sessions.

public static final Key<Integer> EXTENSION_STRENGTH

Strength of the extension post-processing effect

This control allows Camera extension clients to configure the strength of the applied extension effect.

public static final Key<Integer> FLASH_MODE

The desired mode for for the camera device's flash control.

public static final Key<Integer> FLASH_STATE

Current state of the flash unit.

public static final Key<Integer> FLASH_STRENGTH_LEVEL

Flash strength level to be used when manual flash control is active.

public static final Key<Integer> HOT_PIXEL_MODE

Operational mode for hot pixel correction.

public static final Key<Location> JPEG_GPS_LOCATION

A location object to use when generating image GPS metadata.

public static final Key<Integer> JPEG_ORIENTATION

The orientation for a JPEG image.

public static final Key<Byte> JPEG_QUALITY

Compression quality of the final JPEG image.

public static final Key<Byte> JPEG_THUMBNAIL_QUALITY

Compression quality of JPEG thumbnail.

public static final Key<Size> JPEG_THUMBNAIL_SIZE

Resolution of embedded JPEG thumbnail.

public static final Key<Float> LENS_APERTURE

The desired lens aperture size, as a ratio of lens focal length to the effective aperture diameter.

public static final Key<float[]> LENS_DISTORTION

The correction coefficients to correct for this camera device's radial and tangential lens distortion.

public static final Key<Float> LENS_FILTER_DENSITY

The desired setting for the lens neutral density filter(s).

public static final Key<Float> LENS_FOCAL_LENGTH

The desired lens focal length; used for optical zoom.

public static final Key<Float> LENS_FOCUS_DISTANCE

Desired distance to plane of sharpest focus, measured from frontmost surface of the lens.

public static final Key<Pair<FloatFloat>> LENS_FOCUS_RANGE

The range of scene distances that are in sharp focus (depth of field).

public static final Key<float[]> LENS_INTRINSIC_CALIBRATION

The parameters for this camera device's intrinsic calibration.

public static final Key<Integer> LENS_OPTICAL_STABILIZATION_MODE

Sets whether the camera device uses optical image stabilization (OIS) when capturing images.

public static final Key<float[]> LENS_POSE_ROTATION

The orientation of the camera relative to the sensor coordinate system.

public static final Key<float[]> LENS_POSE_TRANSLATION

Position of the camera optical center.

public static final Key<float[]> LENS_RADIAL_DISTORTION

This field was deprecated in API level 28.

This field was inconsistently defined in terms of its normalization. Use android.lens.distortion instead.

public static final Key<Integer> LENS_STATE

Current lens status.

public static final Key<String> LOGICAL_MULTI_CAMERA_ACTIVE_PHYSICAL_ID

String containing the ID of the underlying active physical camera.

public static final Key<Rect> LOGICAL_MULTI_CAMERA_ACTIVE_PHYSICAL_SENSOR_CROP_REGION

The current region of the active physical sensor that will be read out for this capture.

public static final Key<Integer> NOISE_REDUCTION_MODE

Mode of operation for the noise reduction algorithm.

public static final Key<Float> REPROCESS_EFFECTIVE_EXPOSURE_FACTOR

The amount of exposure time increase factor applied to the original output frame by the application processing before sending for reprocessing.

public static final Key<Byte> REQUEST_PIPELINE_DEPTH

Specifies the number of pipeline stages the frame went through from when it was exposed to when the final completed result was available to the framework.

public static final Key<Rect> SCALER_CROP_REGION

The desired region of the sensor to read out for this capture.

public static final Key<Rect> SCALER_RAW_CROP_REGION

The region of the sensor that corresponds to the RAW read out for this capture when the stream use case of a RAW stream is set to CROPPED_RAW.

public static final Key<Integer> SCALER_ROTATE_AND_CROP

Whether a rotation-and-crop operation is applied to processed outputs from the camera.

public static final Key<float[]> SENSOR_DYNAMIC_BLACK_LEVEL

A per-frame dynamic black level offset for each of the color filter arrangement (CFA) mosaic channels.

public static final Key<Integer> SENSOR_DYNAMIC_WHITE_LEVEL

Maximum raw value output by sensor for this frame.

public static final Key<Long> SENSOR_EXPOSURE_TIME

Duration each pixel is exposed to light.

public static final Key<Long> SENSOR_FRAME_DURATION

Duration from start of frame readout to start of next frame readout.

public static final Key<Float> SENSOR_GREEN_SPLIT

The worst-case divergence between Bayer green channels.

public static final Key<Rational[]> SENSOR_NEUTRAL_COLOR_POINT

The estimated camera neutral color in the native sensor colorspace at the time of capture.

public static final Key<Pair[]<DoubleDouble>> SENSOR_NOISE_PROFILE

Noise model coefficients for each CFA mosaic channel.

public static final Key<Integer> SENSOR_PIXEL_MODE

Switches sensor pixel mode between maximum resolution mode and default mode.

public static final Key<Boolean> SENSOR_RAW_BINNING_FACTOR_USED

Whether RAW images requested have their bayer pattern as described by android.sensor.info.binningFactor.

public static final Key<Long> SENSOR_ROLLING_SHUTTER_SKEW

Duration between the start of exposure for the first row of the image sensor, and the start of exposure for one past the last row of the image sensor.

public static final Key<Integer> SENSOR_SENSITIVITY

The amount of gain applied to sensor data before processing.

public static final Key<int[]> SENSOR_TEST_PATTERN_DATA

A pixel [R, G_even, G_odd, B] that supplies the test pattern when android.sensor.testPatternMode is SOLID_COLOR.

public static final Key<Integer> SENSOR_TEST_PATTERN_MODE

When enabled, the sensor sends a test pattern instead of doing a real exposure from the camera.

public static final Key<Long> SENSOR_TIMESTAMP

Time at start of exposure of first row of the image sensor active array, in nanoseconds.

public static final Key<Integer> SHADING_MODE

Quality of lens shading correction applied to the image data.

public static final Key<Face[]> STATISTICS_FACES

List of the faces detected through camera face detection in this capture.

public static final Key<Integer> STATISTICS_FACE_DETECT_MODE

Operating mode for the face detector unit.

public static final Key<Point[]> STATISTICS_HOT_PIXEL_MAP

List of (x, y) coordinates of hot/defective pixels on the sensor.

public static final Key<Boolean> STATISTICS_HOT_PIXEL_MAP_MODE

Operating mode for hot pixel map generation.

public static final Key<LensIntrinsicsSample[]> STATISTICS_LENS_INTRINSICS_SAMPLES

An array of intra-frame lens intrinsic samples.

public static final Key<LensShadingMap> STATISTICS_LENS_SHADING_CORRECTION_MAP

The shading map is a low-resolution floating-point map that lists the coefficients used to correct for vignetting, for each Bayer color channel.

public static final Key<Integer> STATISTICS_LENS_SHADING_MAP_MODE

Whether the camera device will output the lens shading map in output result metadata.

public static final Key<Integer> STATISTICS_OIS_DATA_MODE

A control for selecting whether optical stabilization (OIS) position information is included in output result metadata.

public static final Key<OisSample[]> STATISTICS_OIS_SAMPLES

An array of optical stabilization (OIS) position samples.

public static final Key<Integer> STATISTICS_SCENE_FLICKER

The camera device estimated scene illumination lighting frequency.

public static final Key<TonemapCurve> TONEMAP_CURVE

Tonemapping / contrast / gamma curve to use when android.tonemap.mode is CONTRAST_CURVE.

public static final Key<Float> TONEMAP_GAMMA

Tonemapping curve to use when android.tonemap.mode is GAMMA_VALUE

The tonemap curve will be defined the following formula:

  • OUT = pow(IN, 1.0 / gamma)

where IN and OUT is the input pixel value scaled to range [0.0, 1.0], pow is the power function and gamma is the gamma value specified by this key.

public static final Key<Integer> TONEMAP_MODE

High-level global contrast/gamma/tonemapping control.

public static final Key<Integer> TONEMAP_PRESET_CURVE

Tonemapping curve to use when android.tonemap.mode is PRESET_CURVE

The tonemap curve will be defined by specified standard.

Public methods

<T> T get(Key<T> key)

Get a capture result field value.

String getCameraId()

Get the camera ID of the camera that produced this capture result.

long getFrameNumber()

Get the frame number associated with this result.

List<Key<?>> getKeys()

Returns a list of the keys contained in this map.

CaptureRequest getRequest()

Get the request associated with this result.

int getSequenceId()

The sequence ID for this failure that was returned by the CameraCaptureSession#capture family of functions.

Inherited methods

Fields

BLACK_LEVEL_LOCK

Added in API level 21
public static final Key<Boolean> BLACK_LEVEL_LOCK

Whether black-level compensation is locked to its current values, or is free to vary.

Whether the black level offset was locked for this frame. Should be ON if android.blackLevel.lock was ON in the capture request, unless a change in other capture settings forced the camera device to perform a black level reset.

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

COLOR_CORRECTION_ABERRATION_MODE

Added in API level 21
public static final Key<Integer> COLOR_CORRECTION_ABERRATION_MODE

Mode of operation for the chromatic aberration correction algorithm.

Chromatic (color) aberration is caused by the fact that different wavelengths of light can not focus on the same point after exiting from the lens. This metadata defines the high level control of chromatic aberration correction algorithm, which aims to minimize the chromatic artifacts that may occur along the object boundaries in an image.

FAST/HIGH_QUALITY both mean that camera device determined aberration correction will be applied. HIGH_QUALITY mode indicates that the camera device will use the highest-quality aberration correction algorithms, even if it slows down capture rate. FAST means the camera device will not slow down capture rate when applying aberration correction.

LEGACY devices will always be in FAST mode.

Possible values:

Available values for this device:
android.colorCorrection.availableAberrationModes

This key is available on all devices.

COLOR_CORRECTION_GAINS

Added in API level 21
public static final Key<RggbChannelVector> COLOR_CORRECTION_GAINS

Gains applying to Bayer raw color channels for white-balance.

These per-channel gains are either set by the camera device when the request android.colorCorrection.mode is not TRANSFORM_MATRIX, or directly by the application in the request when the android.colorCorrection.mode is TRANSFORM_MATRIX.

The gains in the result metadata are the gains actually applied by the camera device to the current frame.

The valid range of gains varies on different devices, but gains between [1.0, 3.0] are guaranteed not to be clipped. Even if a given device allows gains below 1.0, this is usually not recommended because this can create color artifacts.

Units: Unitless gain factors

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

COLOR_CORRECTION_MODE

Added in API level 21
public static final Key<Integer> COLOR_CORRECTION_MODE

The mode control selects how the image data is converted from the sensor's native color into linear sRGB color.

When auto-white balance (AWB) is enabled with android.control.awbMode, this control is overridden by the AWB routine. When AWB is disabled, the application controls how the color mapping is performed.

We define the expected processing pipeline below. For consistency across devices, this is always the case with TRANSFORM_MATRIX.

When either FAST or HIGH_QUALITY is used, the camera device may do additional processing but android.colorCorrection.gains and android.colorCorrection.transform will still be provided by the camera device (in the results) and be roughly correct.

Switching to TRANSFORM_MATRIX and using the data provided from FAST or HIGH_QUALITY will yield a picture with the same white point as what was produced by the camera device in the earlier frame.

The expected processing pipeline is as follows:

White balance processing pipeline

The white balance is encoded by two values, a 4-channel white-balance gain vector (applied in the Bayer domain), and a 3x3 color transform matrix (applied after demosaic).

The 4-channel white-balance gains are defined as:

android.colorCorrection.gains = [ R G_even G_odd B ]
 

where G_even is the gain for green pixels on even rows of the output, and G_odd is the gain for green pixels on the odd rows. These may be identical for a given camera device implementation; if the camera device does not support a separate gain for even/odd green channels, it will use the G_even value, and write G_odd equal to G_even in the output result metadata.

The matrices for color transforms are defined as a 9-entry vector:

android.colorCorrection.transform = [ I0 I1 I2 I3 I4 I5 I6 I7 I8 ]
 

which define a transform from input sensor colors, P_in = [ r g b ], to output linear sRGB, P_out = [ r' g' b' ],

with colors as follows:

r' = I0r + I1g + I2b
 g' = I3r + I4g + I5b
 b' = I6r + I7g + I8b
 

Both the input and output value ranges must match. Overflow/underflow values are clipped to fit within the range.

Possible values:

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

COLOR_CORRECTION_TRANSFORM

Added in API level 21
public static final Key<ColorSpaceTransform> COLOR_CORRECTION_TRANSFORM

A color transform matrix to use to transform from sensor RGB color space to output linear sRGB color space.

This matrix is either set by the camera device when the request android.colorCorrection.mode is not TRANSFORM_MATRIX, or directly by the application in the request when the android.colorCorrection.mode is TRANSFORM_MATRIX.

In the latter case, the camera device may round the matrix to account for precision issues; the final rounded matrix should be reported back in this matrix result metadata. The transform should keep the magnitude of the output color values within [0, 1.0] (assuming input color values is within the normalized range [0, 1.0]), or clipping may occur.

The valid range of each matrix element varies on different devices, but values within [-1.5, 3.0] are guaranteed not to be clipped.

Units: Unitless scale factors

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

CONTROL_AE_ANTIBANDING_MODE

Added in API level 21
public static final Key<Integer> CONTROL_AE_ANTIBANDING_MODE

The desired setting for the camera device's auto-exposure algorithm's antibanding compensation.

Some kinds of lighting fixtures, such as some fluorescent lights, flicker at the rate of the power supply frequency (60Hz or 50Hz, depending on country). While this is typically not noticeable to a person, it can be visible to a camera device. If a camera sets its exposure time to the wrong value, the flicker may become visible in the viewfinder as flicker or in a final captured image, as a set of variable-brightness bands across the image.

Therefore, the auto-exposure routines of camera devices include antibanding routines that ensure that the chosen exposure value will not cause such banding. The choice of exposure time depends on the rate of flicker, which the camera device can detect automatically, or the expected rate can be selected by the application using this control.

A given camera device may not support all of the possible options for the antibanding mode. The android.control.aeAvailableAntibandingModes key contains the available modes for a given camera device.

AUTO mode is the default if it is available on given camera device. When AUTO mode is not available, the default will be either 50HZ or 60HZ, and both 50HZ and 60HZ will be available.

If manual exposure control is enabled (by setting android.control.aeMode or android.control.mode to OFF), then this setting has no effect, and the application must ensure it selects exposure times that do not cause banding issues. The android.statistics.sceneFlicker key can assist the application in this.

Possible values:

Available values for this device:

android.control.aeAvailableAntibandingModes

This key is available on all devices.

CONTROL_AE_EXPOSURE_COMPENSATION

Added in API level 21
public static final Key<Integer> CONTROL_AE_EXPOSURE_COMPENSATION

Adjustment to auto-exposure (AE) target image brightness.

The adjustment is measured as a count of steps, with the step size defined by android.control.aeCompensationStep and the allowed range by android.control.aeCompensationRange.

For example, if the exposure value (EV) step is 0.333, '6' will mean an exposure compensation of +2 EV; -3 will mean an exposure compensation of -1 EV. One EV represents a doubling of image brightness. Note that this control will only be effective if android.control.aeMode != OFF. This control will take effect even when android.control.aeLock == true.

In the event of exposure compensation value being changed, camera device may take several frames to reach the newly requested exposure target. During that time, android.control.aeState field will be in the SEARCHING state. Once the new exposure target is reached, android.control.aeState will change from SEARCHING to either CONVERGED, LOCKED (if AE lock is enabled), or FLASH_REQUIRED (if the scene is too dark for still capture).

Units: Compensation steps

Range of valid values:
android.control.aeCompensationRange

This key is available on all devices.

CONTROL_AE_LOCK

Added in API level 21
public static final Key<Boolean> CONTROL_AE_LOCK

Whether auto-exposure (AE) is currently locked to its latest calculated values.

When set to true (ON), the AE algorithm is locked to its latest parameters, and will not change exposure settings until the lock is set to false (OFF).

Note that even when AE is locked, the flash may be fired if the android.control.aeMode is ON_AUTO_FLASH / ON_ALWAYS_FLASH / ON_AUTO_FLASH_REDEYE.

When android.control.aeExposureCompensation is changed, even if the AE lock is ON, the camera device will still adjust its exposure value.

If AE precapture is triggered (see android.control.aePrecaptureTrigger) when AE is already locked, the camera device will not change the exposure time (android.sensor.exposureTime) and sensitivity (android.sensor.sensitivity) parameters. The flash may be fired if the android.control.aeMode is ON_AUTO_FLASH/ON_AUTO_FLASH_REDEYE and the scene is too dark. If the android.control.aeMode is ON_ALWAYS_FLASH, the scene may become overexposed. Similarly, AE precapture trigger CANCEL has no effect when AE is already locked.

When an AE precapture sequence is triggered, AE unlock will not be able to unlock the AE if AE is locked by the camera device internally during precapture metering sequence In other words, submitting requests with AE unlock has no effect for an ongoing precapture metering sequence. Otherwise, the precapture metering sequence will never succeed in a sequence of preview requests where AE lock is always set to false.

Since the camera device has a pipeline of in-flight requests, the settings that get locked do not necessarily correspond to the settings that were present in the latest capture result received from the camera device, since additional captures and AE updates may have occurred even before the result was sent out. If an application is switching between automatic and manual control and wishes to eliminate any flicker during the switch, the following procedure is recommended:

  1. Starting in auto-AE mode:
  2. Lock AE
  3. Wait for the first result to be output that has the AE locked
  4. Copy exposure settings from that result into a request, set the request to manual AE
  5. Submit the capture request, proceed to run manual AE as desired.

See android.control.aeState for AE lock related state transition details.

This key is available on all devices.

CONTROL_AE_MODE

Added in API level 21
public static final Key<Integer> CONTROL_AE_MODE

The desired mode for the camera device's auto-exposure routine.

This control is only effective if android.control.mode is AUTO.

When set to any of the ON modes, the camera device's auto-exposure routine is enabled, overriding the application's selected exposure time, sensor sensitivity, and frame duration (android.sensor.exposureTime, android.sensor.sensitivity, and android.sensor.frameDuration). If one of the FLASH modes is selected, the camera device's flash unit controls are also overridden.

The FLASH modes are only available if the camera device has a flash unit (android.flash.info.available is true).

If flash TORCH mode is desired, this field must be set to ON or OFF, and android.flash.mode set to TORCH.

When set to any of the ON modes, the values chosen by the camera device auto-exposure routine for the overridden fields for a given capture will be available in its CaptureResult.

Possible values:

Available values for this device:
android.control.aeAvailableModes

This key is available on all devices.

CONTROL_AE_PRECAPTURE_TRIGGER

Added in API level 21
public static final Key<Integer> CONTROL_AE_PRECAPTURE_TRIGGER

Whether the camera device will trigger a precapture metering sequence when it processes this request.

This entry is normally set to IDLE, or is not included at all in the request settings. When included and set to START, the camera device will trigger the auto-exposure (AE) precapture metering sequence.

When set to CANCEL, the camera device will cancel any active precapture metering trigger, and return to its initial AE state. If a precapture metering sequence is already completed, and the camera device has implicitly locked the AE for subsequent still capture, the CANCEL trigger will unlock the AE and return to its initial AE state.

The precapture sequence should be triggered before starting a high-quality still capture for final metering decisions to be made, and for firing pre-capture flash pulses to estimate scene brightness and required final capture flash power, when the flash is enabled.

Normally, this entry should be set to START for only a single request, and the application should wait until the sequence completes before starting a new one.

When a precapture metering sequence is finished, the camera device may lock the auto-exposure routine internally to be able to accurately expose the subsequent still capture image (android.control.captureIntent == STILL_CAPTURE). For this case, the AE may not resume normal scan if no subsequent still capture is submitted. To ensure that the AE routine restarts normal scan, the application should submit a request with android.control.aeLock == true, followed by a request with android.control.aeLock == false, if the application decides not to submit a still capture request after the precapture sequence completes. Alternatively, for API level 23 or newer devices, the CANCEL can be used to unlock the camera device internally locked AE if the application doesn't submit a still capture request after the AE precapture trigger. Note that, the CANCEL was added in API level 23, and must not be used in devices that have earlier API levels.

The exact effect of auto-exposure (AE) precapture trigger depends on the current AE mode and state; see android.control.aeState for AE precapture state transition details.

On LEGACY-level devices, the precapture trigger is not supported; capturing a high-resolution JPEG image will automatically trigger a precapture sequence before the high-resolution capture, including potentially firing a pre-capture flash.

Using the precapture trigger and the auto-focus trigger android.control.afTrigger simultaneously is allowed. However, since these triggers often require cooperation between the auto-focus and auto-exposure routines (for example, the may need to be enabled for a focus sweep), the camera device may delay acting on a later trigger until the previous trigger has been fully handled. This may lead to longer intervals between the trigger and changes to android.control.aeState indicating the start of the precapture sequence, for example.

If both the precapture and the auto-focus trigger are activated on the same request, then the camera device will complete them in the optimal order for that device.

Possible values:

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

CONTROL_AE_REGIONS

Added in API level 21
public static final Key<MeteringRectangle[]> CONTROL_AE_REGIONS

List of metering areas to use for auto-exposure adjustment.

Not available if android.control.maxRegionsAe is 0. Otherwise will always be present.

The maximum number of regions supported by the device is determined by the value of android.control.maxRegionsAe.

For devices not supporting android.distortionCorrection.mode control, the coordinate system always follows that of android.sensor.info.activeArraySize, with (0,0) being the top-left pixel in the active pixel array, and (android.sensor.info.activeArraySize.width - 1, android.sensor.info.activeArraySize.height - 1) being the bottom-right pixel in the active pixel array.

For devices supporting android.distortionCorrection.mode control, the coordinate system depends on the mode being set. When the distortion correction mode is OFF, the coordinate system follows android.sensor.info.preCorrectionActiveArraySize, with (0, 0) being the top-left pixel of the pre-correction active array, and (android.sensor.info.preCorrectionActiveArraySize.width - 1, android.sensor.info.preCorrectionActiveArraySize.height - 1) being the bottom-right pixel in the pre-correction active pixel array. When the distortion correction mode is not OFF, the coordinate system follows android.sensor.info.activeArraySize, with (0, 0) being the top-left pixel of the active array, and (android.sensor.info.activeArraySize.width - 1, android.sensor.info.activeArraySize.height - 1) being the bottom-right pixel in the active pixel array.

The weight must be within [0, 1000], and represents a weight for every pixel in the area. This means that a large metering area with the same weight as a smaller area will have more effect in the metering result. Metering areas can partially overlap and the camera device will add the weights in the overlap region.

The weights are relative to weights of other exposure metering regions, so if only one region is used, all non-zero weights will have the same effect. A region with 0 weight is ignored.

If all regions have 0 weight, then no specific metering area needs to be used by the camera device.

If the metering region is outside the used android.scaler.cropRegion returned in capture result metadata, the camera device will ignore the sections outside the crop region and output only the intersection rectangle as the metering region in the result metadata. If the region is entirely outside the crop region, it will be ignored and not reported in the result metadata.

When setting the AE metering regions, the application must consider the additional crop resulted from the aspect ratio differences between the preview stream and android.scaler.cropRegion. For example, if the android.scaler.cropRegion is the full active array size with 4:3 aspect ratio, and the preview stream is 16:9, the boundary of AE regions will be [0, y_crop] and [active_width, active_height - 2 * y_crop] rather than [0, 0] and [active_width, active_height], where y_crop is the additional crop due to aspect ratio mismatch.

Starting from API level 30, the coordinate system of activeArraySize or preCorrectionActiveArraySize is used to represent post-zoomRatio field of view, not pre-zoom field of view. This means that the same aeRegions values at different android.control.zoomRatio represent different parts of the scene. The aeRegions coordinates are relative to the activeArray/preCorrectionActiveArray representing the zoomed field of view. If android.control.zoomRatio is set to 1.0 (default), the same aeRegions at different android.scaler.cropRegion still represent the same parts of the scene as they do before. See android.control.zoomRatio for details. Whether to use activeArraySize or preCorrectionActiveArraySize still depends on distortion correction mode.

For camera devices with the CameraMetadata.REQUEST_AVAILABLE_CAPABILITIES_ULTRA_HIGH_RESOLUTION_SENSOR capability or devices where CameraCharacteristics#getAvailableCaptureRequestKeys lists CaptureRequest.SENSOR_PIXEL_MODE} android.sensor.info.activeArraySizeMaximumResolution / android.sensor.info.preCorrectionActiveArraySizeMaximumResolution must be used as the coordinate system for requests where android.sensor.pixelMode is set to CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION.

Units: Pixel coordinates within android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

Range of valid values:
Coordinates must be between [(0,0), (width, height)) of android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

Optional - The value for this key may be null on some devices.

CONTROL_AE_STATE

Added in API level 21
public static final Key<Integer> CONTROL_AE_STATE

Current state of the auto-exposure (AE) algorithm.

Switching between or enabling AE modes (android.control.aeMode) always resets the AE state to INACTIVE. Similarly, switching between android.control.mode, or android.control.sceneMode if android.control.mode == USE_SCENE_MODE resets all the algorithm states to INACTIVE.

The camera device can do several state transitions between two results, if it is allowed by the state transition table. For example: INACTIVE may never actually be seen in a result.

The state in the result is the state for this image (in sync with this image): if AE state becomes CONVERGED, then the image data associated with this result should be good to use.

Below are state transition tables for different AE modes.

State Transition Cause New State Notes
INACTIVE INACTIVE Camera device auto exposure algorithm is disabled

When android.control.aeMode is AE_MODE_ON*:

State Transition Cause New State Notes
INACTIVE Camera device initiates AE scan SEARCHING Values changing
INACTIVE android.control.aeLock is ON LOCKED Values locked
SEARCHING Camera device finishes AE scan CONVERGED Good values, not changing
SEARCHING Camera device finishes AE scan FLASH_REQUIRED Converged but too dark w/o flash
SEARCHING android.control.aeLock is ON LOCKED Values locked
CONVERGED Camera device initiates AE scan SEARCHING Values changing
CONVERGED android.control.aeLock is ON LOCKED Values locked
FLASH_REQUIRED Camera device initiates AE scan SEARCHING Values changing
FLASH_REQUIRED android.control.aeLock is ON LOCKED Values locked
LOCKED android.control.aeLock is OFF SEARCHING Values not good after unlock
LOCKED android.control.aeLock is OFF CONVERGED Values good after unlock
LOCKED android.control.aeLock is OFF FLASH_REQUIRED Exposure good, but too dark
PRECAPTURE Sequence done. android.control.aeLock is OFF CONVERGED Ready for high-quality capture
PRECAPTURE Sequence done. android.control.aeLock is ON LOCKED Ready for high-quality capture
LOCKED aeLock is ON and aePrecaptureTrigger is START LOCKED Precapture trigger is ignored when AE is already locked
LOCKED aeLock is ON and aePrecaptureTrigger is CANCEL LOCKED Precapture trigger is ignored when AE is already locked
Any state (excluding LOCKED) android.control.aePrecaptureTrigger is START PRECAPTURE Start AE precapture metering sequence
Any state (excluding LOCKED) android.control.aePrecaptureTrigger is CANCEL INACTIVE Currently active precapture metering sequence is canceled

If the camera device supports AE external flash mode (ON_EXTERNAL_FLASH is included in android.control.aeAvailableModes), android.control.aeState must be FLASH_REQUIRED after the camera device finishes AE scan and it's too dark without flash.

For the above table, the camera device may skip reporting any state changes that happen without application intervention (i.e. mode switch, trigger, locking). Any state that can be skipped in that manner is called a transient state.

For example, for above AE modes (AE_MODE_ON*), in addition to the state transitions listed in above table, it is also legal for the camera device to skip one or more transient states between two results. See below table for examples:

State Transition Cause New State Notes
INACTIVE Camera device finished AE scan CONVERGED Values are already good, transient states are skipped by camera device.
Any state (excluding LOCKED) android.control.aePrecaptureTrigger is START, sequence done FLASH_REQUIRED Converged but too dark w/o flash after a precapture sequence, transient states are skipped by camera device.
Any state (excluding LOCKED) android.control.aePrecaptureTrigger is START, sequence done CONVERGED Converged after a precapture sequence, transient states are skipped by camera device.
Any state (excluding LOCKED) android.control.aePrecaptureTrigger is CANCEL, converged FLASH_REQUIRED Converged but too dark w/o flash after a precapture sequence is canceled, transient states are skipped by camera device.
Any state (excluding LOCKED) android.control.aePrecaptureTrigger is CANCEL, converged CONVERGED Converged after a precapture sequences canceled, transient states are skipped by camera device.
CONVERGED Camera device finished AE scan FLASH_REQUIRED Converged but too dark w/o flash after a new scan, transient states are skipped by camera device.
FLASH_REQUIRED Camera device finished AE scan CONVERGED Converged after a new scan, transient states are skipped by camera device.

Possible values:

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

CONTROL_AE_TARGET_FPS_RANGE

Added in API level 21
public static final Key<Range<Integer>> CONTROL_AE_TARGET_FPS_RANGE

Range over which the auto-exposure routine can adjust the capture frame rate to maintain good exposure.

Only constrains auto-exposure (AE) algorithm, not manual control of android.sensor.exposureTime and android.sensor.frameDuration.

Note that the actual achievable max framerate also depends on the minimum frame duration of the output streams. The max frame rate will be min(aeTargetFpsRange.maxFps, 1 / max(individual stream min durations)). For example, if the application sets this key to {60, 60}, but the maximum minFrameDuration among all configured streams is 33ms, the maximum framerate won't be 60fps, but will be 30fps.

To start a CaptureSession with a target FPS range different from the capture request template's default value, the application is strongly recommended to call SessionConfiguration.setSessionParameters(CaptureRequest) with the target fps range before creating the capture session. The aeTargetFpsRange is typically a session parameter. Specifying it at session creation time helps avoid session reconfiguration delays in cases like 60fps or high speed recording.

Units: Frames per second (FPS)

Range of valid values:
Any of the entries in android.control.aeAvailableTargetFpsRanges

This key is available on all devices.

CONTROL_AF_MODE

Added in API level 21
public static final Key<Integer> CONTROL_AF_MODE

Whether auto-focus (AF) is currently enabled, and what mode it is set to.

Only effective if android.control.mode = AUTO and the lens is not fixed focus (i.e. android.lens.info.minimumFocusDistance > 0). Also note that when android.control.aeMode is OFF, the behavior of AF is device dependent. It is recommended to lock AF by using android.control.afTrigger before setting android.control.aeMode to OFF, or set AF mode to OFF when AE is OFF.

If the lens is controlled by the camera device auto-focus algorithm, the camera device will report the current AF status in android.control.afState in result metadata.

Possible values:

Available values for this device:
android.control.afAvailableModes

This key is available on all devices.

CONTROL_AF_REGIONS

Added in API level 21
public static final Key<MeteringRectangle[]> CONTROL_AF_REGIONS

List of metering areas to use for auto-focus.

Not available if android.control.maxRegionsAf is 0. Otherwise will always be present.

The maximum number of focus areas supported by the device is determined by the value of android.control.maxRegionsAf.

For devices not supporting android.distortionCorrection.mode control, the coordinate system always follows that of android.sensor.info.activeArraySize, with (0,0) being the top-left pixel in the active pixel array, and (android.sensor.info.activeArraySize.width - 1, android.sensor.info.activeArraySize.height - 1) being the bottom-right pixel in the active pixel array.

For devices supporting android.distortionCorrection.mode control, the coordinate system depends on the mode being set. When the distortion correction mode is OFF, the coordinate system follows android.sensor.info.preCorrectionActiveArraySize, with (0, 0) being the top-left pixel of the pre-correction active array, and (android.sensor.info.preCorrectionActiveArraySize.width - 1, android.sensor.info.preCorrectionActiveArraySize.height - 1) being the bottom-right pixel in the pre-correction active pixel array. When the distortion correction mode is not OFF, the coordinate system follows android.sensor.info.activeArraySize, with (0, 0) being the top-left pixel of the active array, and (android.sensor.info.activeArraySize.width - 1, android.sensor.info.activeArraySize.height - 1) being the bottom-right pixel in the active pixel array.

The weight must be within [0, 1000], and represents a weight for every pixel in the area. This means that a large metering area with the same weight as a smaller area will have more effect in the metering result. Metering areas can partially overlap and the camera device will add the weights in the overlap region.

The weights are relative to weights of other metering regions, so if only one region is used, all non-zero weights will have the same effect. A region with 0 weight is ignored.

If all regions have 0 weight, then no specific metering area needs to be used by the camera device. The capture result will either be a zero weight region as well, or the region selected by the camera device as the focus area of interest.

If the metering region is outside the used android.scaler.cropRegion returned in capture result metadata, the camera device will ignore the sections outside the crop region and output only the intersection rectangle as the metering region in the result metadata. If the region is entirely outside the crop region, it will be ignored and not reported in the result metadata.

When setting the AF metering regions, the application must consider the additional crop resulted from the aspect ratio differences between the preview stream and android.scaler.cropRegion. For example, if the android.scaler.cropRegion is the full active array size with 4:3 aspect ratio, and the preview stream is 16:9, the boundary of AF regions will be [0, y_crop] and [active_width, active_height - 2 * y_crop] rather than [0, 0] and [active_width, active_height], where y_crop is the additional crop due to aspect ratio mismatch.

Starting from API level 30, the coordinate system of activeArraySize or preCorrectionActiveArraySize is used to represent post-zoomRatio field of view, not pre-zoom field of view. This means that the same afRegions values at different android.control.zoomRatio represent different parts of the scene. The afRegions coordinates are relative to the activeArray/preCorrectionActiveArray representing the zoomed field of view. If android.control.zoomRatio is set to 1.0 (default), the same afRegions at different android.scaler.cropRegion still represent the same parts of the scene as they do before. See android.control.zoomRatio for details. Whether to use activeArraySize or preCorrectionActiveArraySize still depends on distortion correction mode.

For camera devices with the CameraMetadata.REQUEST_AVAILABLE_CAPABILITIES_ULTRA_HIGH_RESOLUTION_SENSOR capability or devices where CameraCharacteristics#getAvailableCaptureRequestKeys lists CaptureRequest.SENSOR_PIXEL_MODE}, android.sensor.info.activeArraySizeMaximumResolution / android.sensor.info.preCorrectionActiveArraySizeMaximumResolution must be used as the coordinate system for requests where android.sensor.pixelMode is set to CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION.

Units: Pixel coordinates within android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

Range of valid values:
Coordinates must be between [(0,0), (width, height)) of android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

Optional - The value for this key may be null on some devices.

CONTROL_AF_SCENE_CHANGE

Added in API level 28
public static final Key<Integer> CONTROL_AF_SCENE_CHANGE

Whether a significant scene change is detected within the currently-set AF region(s).

When the camera focus routine detects a change in the scene it is looking at, such as a large shift in camera viewpoint, significant motion in the scene, or a significant illumination change, this value will be set to DETECTED for a single capture result. Otherwise the value will be NOT_DETECTED. The threshold for detection is similar to what would trigger a new passive focus scan to begin in CONTINUOUS autofocus modes.

This key will be available if the camera device advertises this key via CameraCharacteristics.getAvailableCaptureResultKeys().

Possible values:

Optional - The value for this key may be null on some devices.

CONTROL_AF_STATE

Added in API level 21
public static final Key<Integer> CONTROL_AF_STATE

Current state of auto-focus (AF) algorithm.

Switching between or enabling AF modes (android.control.afMode) always resets the AF state to INACTIVE. Similarly, switching between android.control.mode, or android.control.sceneMode if android.control.mode == USE_SCENE_MODE resets all the algorithm states to INACTIVE.

The camera device can do several state transitions between two results, if it is allowed by the state transition table. For example: INACTIVE may never actually be seen in a result.

The state in the result is the state for this image (in sync with this image): if AF state becomes FOCUSED, then the image data associated with this result should be sharp.

Below are state transition tables for different AF modes.

When android.control.afMode is AF_MODE_OFF or AF_MODE_EDOF:

State Transition Cause New State Notes
INACTIVE INACTIVE Never changes

When android.control.afMode is AF_MODE_AUTO or AF_MODE_MACRO:

State Transition Cause New State Notes
INACTIVE AF_TRIGGER ACTIVE_SCAN Start AF sweep, Lens now moving
ACTIVE_SCAN AF sweep done FOCUSED_LOCKED Focused, Lens now locked
ACTIVE_SCAN AF sweep done NOT_FOCUSED_LOCKED Not focused, Lens now locked
ACTIVE_SCAN AF_CANCEL INACTIVE Cancel/reset AF, Lens now locked
FOCUSED_LOCKED AF_CANCEL INACTIVE Cancel/reset AF
FOCUSED_LOCKED AF_TRIGGER ACTIVE_SCAN Start new sweep, Lens now moving
NOT_FOCUSED_LOCKED AF_CANCEL INACTIVE Cancel/reset AF
NOT_FOCUSED_LOCKED AF_TRIGGER ACTIVE_SCAN Start new sweep, Lens now moving
Any state Mode change INACTIVE

For the above table, the camera device may skip reporting any state changes that happen without application intervention (i.e. mode switch, trigger, locking). Any state that can be skipped in that manner is called a transient state.

For example, for these AF modes (AF_MODE_AUTO and AF_MODE_MACRO), in addition to the state transitions listed in above table, it is also legal for the camera device to skip one or more transient states between two results. See below table for examples:

State Transition Cause New State Notes
INACTIVE AF_TRIGGER FOCUSED_LOCKED Focus is already good or good after a scan, lens is now locked.
INACTIVE AF_TRIGGER NOT_FOCUSED_LOCKED Focus failed after a scan, lens is now locked.
FOCUSED_LOCKED AF_TRIGGER FOCUSED_LOCKED Focus is already good or good after a scan, lens is now locked.
NOT_FOCUSED_LOCKED AF_TRIGGER FOCUSED_LOCKED Focus is good after a scan, lens is not locked.

When android.control.afMode is AF_MODE_CONTINUOUS_VIDEO:

State Transition Cause New State Notes
INACTIVE Camera device initiates new scan PASSIVE_SCAN Start AF scan, Lens now moving
INACTIVE AF_TRIGGER NOT_FOCUSED_LOCKED AF state query, Lens now locked
PASSIVE_SCAN Camera device completes current scan PASSIVE_FOCUSED End AF scan, Lens now locked
PASSIVE_SCAN Camera device fails current scan PASSIVE_UNFOCUSED End AF scan, Lens now locked
PASSIVE_SCAN AF_TRIGGER FOCUSED_LOCKED Immediate transition, if focus is good. Lens now locked
PASSIVE_SCAN AF_TRIGGER NOT_FOCUSED_LOCKED Immediate transition, if focus is bad. Lens now locked
PASSIVE_SCAN AF_CANCEL INACTIVE Reset lens position, Lens now locked
PASSIVE_FOCUSED Camera device initiates new scan PASSIVE_SCAN Start AF scan, Lens now moving
PASSIVE_UNFOCUSED Camera device initiates new scan PASSIVE_SCAN Start AF scan, Lens now moving
PASSIVE_FOCUSED AF_TRIGGER FOCUSED_LOCKED Immediate transition, lens now locked
PASSIVE_UNFOCUSED AF_TRIGGER NOT_FOCUSED_LOCKED Immediate transition, lens now locked
FOCUSED_LOCKED AF_TRIGGER FOCUSED_LOCKED No effect
FOCUSED_LOCKED AF_CANCEL INACTIVE Restart AF scan
NOT_FOCUSED_LOCKED AF_TRIGGER NOT_FOCUSED_LOCKED No effect
NOT_FOCUSED_LOCKED AF_CANCEL INACTIVE Restart AF scan

When android.control.afMode is AF_MODE_CONTINUOUS_PICTURE:

State Transition Cause New State Notes
INACTIVE Camera device initiates new scan PASSIVE_SCAN Start AF scan, Lens now moving
INACTIVE AF_TRIGGER NOT_FOCUSED_LOCKED AF state query, Lens now locked
PASSIVE_SCAN Camera device completes current scan PASSIVE_FOCUSED End AF scan, Lens now locked
PASSIVE_SCAN Camera device fails current scan PASSIVE_UNFOCUSED End AF scan, Lens now locked
PASSIVE_SCAN AF_TRIGGER FOCUSED_LOCKED Eventual transition once the focus is good. Lens now locked
PASSIVE_SCAN AF_TRIGGER NOT_FOCUSED_LOCKED Eventual transition if cannot find focus. Lens now locked
PASSIVE_SCAN AF_CANCEL INACTIVE Reset lens position, Lens now locked
PASSIVE_FOCUSED Camera device initiates new scan PASSIVE_SCAN Start AF scan, Lens now moving
PASSIVE_UNFOCUSED Camera device initiates new scan PASSIVE_SCAN Start AF scan, Lens now moving
PASSIVE_FOCUSED AF_TRIGGER FOCUSED_LOCKED Immediate trans. Lens now locked
PASSIVE_UNFOCUSED AF_TRIGGER NOT_FOCUSED_LOCKED Immediate trans. Lens now locked
FOCUSED_LOCKED AF_TRIGGER FOCUSED_LOCKED No effect
FOCUSED_LOCKED AF_CANCEL INACTIVE Restart AF scan
NOT_FOCUSED_LOCKED AF_TRIGGER NOT_FOCUSED_LOCKED No effect
NOT_FOCUSED_LOCKED AF_CANCEL INACTIVE Restart AF scan

When switch between AF_MODE_CONTINUOUS_* (CAF modes) and AF_MODE_AUTO/AF_MODE_MACRO (AUTO modes), the initial INACTIVE or PASSIVE_SCAN states may be skipped by the camera device. When a trigger is included in a mode switch request, the trigger will be evaluated in the context of the new mode in the request. See below table for examples:

State Transition Cause New State Notes
any state CAF-->AUTO mode switch INACTIVE Mode switch without trigger, initial state must be INACTIVE
any state CAF-->AUTO mode switch with AF_TRIGGER trigger-reachable states from INACTIVE Mode switch with trigger, INACTIVE is skipped
any state AUTO-->CAF mode switch passively reachable states from INACTIVE Mode switch without trigger, passive transient state is skipped

Possible values:

This key is available on all devices.

CONTROL_AF_TRIGGER

Added in API level 21
public static final Key<Integer> CONTROL_AF_TRIGGER

Whether the camera device will trigger autofocus for this request.

This entry is normally set to IDLE, or is not included at all in the request settings.

When included and set to START, the camera device will trigger the autofocus algorithm. If autofocus is disabled, this trigger has no effect.

When set to CANCEL, the camera device will cancel any active trigger, and return to its initial AF state.

Generally, applications should set this entry to START or CANCEL for only a single capture, and then return it to IDLE (or not set at all). Specifying START for multiple captures in a row means restarting the AF operation over and over again.

See android.control.afState for what the trigger means for each AF mode.

Using the autofocus trigger and the precapture trigger android.control.aePrecaptureTrigger simultaneously is allowed. However, since these triggers often require cooperation between the auto-focus and auto-exposure routines (for example, the may need to be enabled for a focus sweep), the camera device may delay acting on a later trigger until the previous trigger has been fully handled. This may lead to longer intervals between the trigger and changes to android.control.afState, for example.

Possible values:

This key is available on all devices.

CONTROL_AUTOFRAMING

Added in API level 34
public static final Key<Integer> CONTROL_AUTOFRAMING

Automatic crop, pan and zoom to keep objects in the center of the frame.

Auto-framing is a special mode provided by the camera device to dynamically crop, zoom or pan the camera feed to try to ensure that the people in a scene occupy a reasonable portion of the viewport. It is primarily designed to support video calling in situations where the user isn't directly in front of the device, especially for wide-angle cameras. android.scaler.cropRegion and android.control.zoomRatio in CaptureResult will be used to denote the coordinates of the auto-framed region. Zoom and video stabilization controls are disabled when auto-framing is enabled. The 3A regions must map the screen coordinates into the scaler crop returned from the capture result instead of using the active array sensor.

Possible values:

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

CONTROL_AUTOFRAMING_STATE

Added in API level 34
public static final Key<Integer> CONTROL_AUTOFRAMING_STATE

Current state of auto-framing.

When the camera doesn't have auto-framing available (i.e android.control.autoframingAvailable == false) or it is not enabled (i.e android.control.autoframing == OFF), the state will always be INACTIVE. Other states indicate the current auto-framing state:

  • When android.control.autoframing is set to ON, auto-framing will take place. While the frame is aligning itself to center the object (doing things like zooming in, zooming out or pan), the state will be FRAMING.
  • When field of view is not being adjusted anymore and has reached a stable state, the state will be CONVERGED.

Possible values:

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

CONTROL_AWB_LOCK

Added in API level 21
public static final Key<Boolean> CONTROL_AWB_LOCK

Whether auto-white balance (AWB) is currently locked to its latest calculated values.

When set to true (ON), the AWB algorithm is locked to its latest parameters, and will not change color balance settings until the lock is set to false (OFF).

Since the camera device has a pipeline of in-flight requests, the settings that get locked do not necessarily correspond to the settings that were present in the latest capture result received from the camera device, since additional captures and AWB updates may have occurred even before the result was sent out. If an application is switching between automatic and manual control and wishes to eliminate any flicker during the switch, the following procedure is recommended:

  1. Starting in auto-AWB mode:
  2. Lock AWB
  3. Wait for the first result to be output that has the AWB locked
  4. Copy AWB settings from that result into a request, set the request to manual AWB
  5. Submit the capture request, proceed to run manual AWB as desired.

Note that AWB lock is only meaningful when android.control.awbMode is in the AUTO mode; in other modes, AWB is already fixed to a specific setting.

Some LEGACY devices may not support ON; the value is then overridden to OFF.

This key is available on all devices.

CONTROL_AWB_MODE

Added in API level 21
public static final Key<Integer> CONTROL_AWB_MODE

Whether auto-white balance (AWB) is currently setting the color transform fields, and what its illumination target is.

This control is only effective if android.control.mode is AUTO.

When set to the AUTO mode, the camera device's auto-white balance routine is enabled, overriding the application's selected android.colorCorrection.transform, android.colorCorrection.gains and android.colorCorrection.mode. Note that when android.control.aeMode is OFF, the behavior of AWB is device dependent. It is recommended to also set AWB mode to OFF or lock AWB by using android.control.awbLock before setting AE mode to OFF.

When set to the OFF mode, the camera device's auto-white balance routine is disabled. The application manually controls the white balance by android.colorCorrection.transform, android.colorCorrection.gains and android.colorCorrection.mode.

When set to any other modes, the camera device's auto-white balance routine is disabled. The camera device uses each particular illumination target for white balance adjustment. The application's values for android.colorCorrection.transform, android.colorCorrection.gains and android.colorCorrection.mode are ignored.

Possible values:

Available values for this device:
android.control.awbAvailableModes

This key is available on all devices.

CONTROL_AWB_REGIONS

Added in API level 21
public static final Key<MeteringRectangle[]> CONTROL_AWB_REGIONS

List of metering areas to use for auto-white-balance illuminant estimation.

Not available if android.control.maxRegionsAwb is 0. Otherwise will always be present.

The maximum number of regions supported by the device is determined by the value of android.control.maxRegionsAwb.

For devices not supporting android.distortionCorrection.mode control, the coordinate system always follows that of android.sensor.info.activeArraySize, with (0,0) being the top-left pixel in the active pixel array, and (android.sensor.info.activeArraySize.width - 1, android.sensor.info.activeArraySize.height - 1) being the bottom-right pixel in the active pixel array.

For devices supporting android.distortionCorrection.mode control, the coordinate system depends on the mode being set. When the distortion correction mode is OFF, the coordinate system follows android.sensor.info.preCorrectionActiveArraySize, with (0, 0) being the top-left pixel of the pre-correction active array, and (android.sensor.info.preCorrectionActiveArraySize.width - 1, android.sensor.info.preCorrectionActiveArraySize.height - 1) being the bottom-right pixel in the pre-correction active pixel array. When the distortion correction mode is not OFF, the coordinate system follows android.sensor.info.activeArraySize, with (0, 0) being the top-left pixel of the active array, and (android.sensor.info.activeArraySize.width - 1, android.sensor.info.activeArraySize.height - 1) being the bottom-right pixel in the active pixel array.

The weight must range from 0 to 1000, and represents a weight for every pixel in the area. This means that a large metering area with the same weight as a smaller area will have more effect in the metering result. Metering areas can partially overlap and the camera device will add the weights in the overlap region.

The weights are relative to weights of other white balance metering regions, so if only one region is used, all non-zero weights will have the same effect. A region with 0 weight is ignored.

If all regions have 0 weight, then no specific metering area needs to be used by the camera device.

If the metering region is outside the used android.scaler.cropRegion returned in capture result metadata, the camera device will ignore the sections outside the crop region and output only the intersection rectangle as the metering region in the result metadata. If the region is entirely outside the crop region, it will be ignored and not reported in the result metadata.

When setting the AWB metering regions, the application must consider the additional crop resulted from the aspect ratio differences between the preview stream and android.scaler.cropRegion. For example, if the android.scaler.cropRegion is the full active array size with 4:3 aspect ratio, and the preview stream is 16:9, the boundary of AWB regions will be [0, y_crop] and [active_width, active_height - 2 * y_crop] rather than [0, 0] and [active_width, active_height], where y_crop is the additional crop due to aspect ratio mismatch.

Starting from API level 30, the coordinate system of activeArraySize or preCorrectionActiveArraySize is used to represent post-zoomRatio field of view, not pre-zoom field of view. This means that the same awbRegions values at different android.control.zoomRatio represent different parts of the scene. The awbRegions coordinates are relative to the activeArray/preCorrectionActiveArray representing the zoomed field of view. If android.control.zoomRatio is set to 1.0 (default), the same awbRegions at different android.scaler.cropRegion still represent the same parts of the scene as they do before. See android.control.zoomRatio for details. Whether to use activeArraySize or preCorrectionActiveArraySize still depends on distortion correction mode.

For camera devices with the CameraMetadata.REQUEST_AVAILABLE_CAPABILITIES_ULTRA_HIGH_RESOLUTION_SENSOR capability or devices where CameraCharacteristics#getAvailableCaptureRequestKeys lists CaptureRequest.SENSOR_PIXEL_MODE}, android.sensor.info.activeArraySizeMaximumResolution / android.sensor.info.preCorrectionActiveArraySizeMaximumResolution must be used as the coordinate system for requests where android.sensor.pixelMode is set to CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION.

Units: Pixel coordinates within android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

Range of valid values:
Coordinates must be between [(0,0), (width, height)) of android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

Optional - The value for this key may be null on some devices.

CONTROL_AWB_STATE

Added in API level 21
public static final Key<Integer> CONTROL_AWB_STATE

Current state of auto-white balance (AWB) algorithm.

Switching between or enabling AWB modes (android.control.awbMode) always resets the AWB state to INACTIVE. Similarly, switching between android.control.mode, or android.control.sceneMode if android.control.mode == USE_SCENE_MODE resets all the algorithm states to INACTIVE.

The camera device can do several state transitions between two results, if it is allowed by the state transition table. So INACTIVE may never actually be seen in a result.

The state in the result is the state for this image (in sync with this image): if AWB state becomes CONVERGED, then the image data associated with this result should be good to use.

Below are state transition tables for different AWB modes.

When android.control.awbMode != AWB_MODE_AUTO:

State Transition Cause New State Notes
INACTIVE INACTIVE Camera device auto white balance algorithm is disabled

When android.control.awbMode is AWB_MODE_AUTO:

State Transition Cause New State Notes
INACTIVE Camera device initiates AWB scan SEARCHING Values changing
INACTIVE android.control.awbLock is ON LOCKED Values locked
SEARCHING Camera device finishes AWB scan CONVERGED Good values, not changing
SEARCHING android.control.awbLock is ON LOCKED Values locked
CONVERGED Camera device initiates AWB scan SEARCHING Values changing
CONVERGED android.control.awbLock is ON LOCKED Values locked
LOCKED android.control.awbLock is OFF SEARCHING Values not good after unlock

For the above table, the camera device may skip reporting any state changes that happen without application intervention (i.e. mode switch, trigger, locking). Any state that can be skipped in that manner is called a transient state.

For example, for this AWB mode (AWB_MODE_AUTO), in addition to the state transitions listed in above table, it is also legal for the camera device to skip one or more transient states between two results. See below table for examples:

State Transition Cause New State Notes
INACTIVE Camera device finished AWB scan CONVERGED Values are already good, transient states are skipped by camera device.
LOCKED android.control.awbLock is OFF CONVERGED Values good after unlock, transient states are skipped by camera device.

Possible values:

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

CONTROL_CAPTURE_INTENT

Added in API level 21
public static final Key<Integer> CONTROL_CAPTURE_INTENT

Information to the camera device 3A (auto-exposure, auto-focus, auto-white balance) routines about the purpose of this capture, to help the camera device to decide optimal 3A strategy.

This control (except for MANUAL) is only effective if android.control.mode != OFF and any 3A routine is active.

All intents are supported by all devices, except that:

Possible values:

This key is available on all devices.

CONTROL_EFFECT_MODE

Added in API level 21
public static final Key<Integer> CONTROL_EFFECT_MODE

A special color effect to apply.

When this mode is set, a color effect will be applied to images produced by the camera device. The interpretation and implementation of these color effects is left to the implementor of the camera device, and should not be depended on to be consistent (or present) across all devices.

Possible values:

Available values for this device:
android.control.availableEffects

This key is available on all devices.

CONTROL_ENABLE_ZSL

Added in API level 26
public static final Key<Boolean> CONTROL_ENABLE_ZSL

Allow camera device to enable zero-shutter-lag mode for requests with android.control.captureIntent == STILL_CAPTURE.

If enableZsl is true, the camera device may enable zero-shutter-lag mode for requests with STILL_CAPTURE capture intent. The camera device may use images captured in the past to produce output images for a zero-shutter-lag request. The result metadata including the android.sensor.timestamp reflects the source frames used to produce output images. Therefore, the contents of the output images and the result metadata may be out of order compared to previous regular requests. enableZsl does not affect requests with other capture intents.

For example, when requests are submitted in the following order: Request A: enableZsl is ON, android.control.captureIntent is PREVIEW Request B: enableZsl is ON, android.control.captureIntent is STILL_CAPTURE

The output images for request B may have contents captured before the output images for request A, and the result metadata for request B may be older than the result metadata for request A.

Note that when enableZsl is true, it is not guaranteed to get output images captured in the past for requests with STILL_CAPTURE capture intent.

For applications targeting SDK versions O and newer, the value of enableZsl in TEMPLATE_STILL_CAPTURE template may be true. The value in other templates is always false if present.

For applications targeting SDK versions older than O, the value of enableZsl in all capture templates is always false if present.

For application-operated ZSL, use CAMERA3_TEMPLATE_ZERO_SHUTTER_LAG template.

Optional - The value for this key may be null on some devices.

CONTROL_EXTENDED_SCENE_MODE

Added in API level 30
public static final Key<Integer> CONTROL_EXTENDED_SCENE_MODE

Whether extended scene mode is enabled for a particular capture request.

With bokeh mode, the camera device may blur out the parts of scene that are not in focus, creating a bokeh (or shallow depth of field) effect for people or objects.

When set to BOKEH_STILL_CAPTURE mode with STILL_CAPTURE capture intent, due to the extra processing needed for high quality bokeh effect, the stall may be longer than when capture intent is not STILL_CAPTURE.

When set to BOKEH_STILL_CAPTURE mode with PREVIEW capture intent,

When set to BOKEH_CONTINUOUS mode, configured streams dimension should not exceed this mode's maximum streaming dimension in order to have bokeh effect applied. Bokeh effect may not be available for streams larger than the maximum streaming dimension.

Switching between different extended scene modes may involve reconfiguration of the camera pipeline, resulting in long latency. The application should check this key against the available session keys queried via CameraCharacteristics.getAvailableSessionKeys().

For a logical multi-camera, bokeh may be implemented by stereo vision from sub-cameras with different field of view. As a result, when bokeh mode is enabled, the camera device may override android.scaler.cropRegion or android.control.zoomRatio, and the field of view may be smaller than when bokeh mode is off.

Possible values:

Optional - The value for this key may be null on some devices.

CONTROL_LOW_LIGHT_BOOST_STATE

Added in API level 35
public static final Key<Integer> CONTROL_LOW_LIGHT_BOOST_STATE

Current state of the low light boost AE mode.

When low light boost is enabled by setting the AE mode to 'ON_LOW_LIGHT_BOOST_BRIGHTNESS_PRIORITY', it can dynamically apply a low light boost when the light level threshold is exceeded.

This state indicates when low light boost is 'ACTIVE' and applied. Similarly, it can indicate when it is not being applied by returning 'INACTIVE'.

This key will be absent from the CaptureResult if AE mode is not set to 'ON_LOW_LIGHT_BOOST_BRIGHTNESS_PRIORITY.

The default value will always be 'INACTIVE'.

Possible values:

Optional - The value for this key may be null on some devices.

CONTROL_MODE

Added in API level 21
public static final Key<Integer> CONTROL_MODE

Overall mode of 3A (auto-exposure, auto-white-balance, auto-focus) control routines.

This is a top-level 3A control switch. When set to OFF, all 3A control by the camera device is disabled. The application must set the fields for capture parameters itself.

When set to AUTO, the individual algorithm controls in android.control.* are in effect, such as android.control.afMode.

When set to USE_SCENE_MODE or USE_EXTENDED_SCENE_MODE, the individual controls in android.control.* are mostly disabled, and the camera device implements one of the scene mode or extended scene mode settings (such as ACTION, SUNSET, PARTY, or BOKEH) as it wishes. The camera device scene mode 3A settings are provided by capture results.

When set to OFF_KEEP_STATE, it is similar to OFF mode, the only difference is that this frame will not be used by camera device background 3A statistics update, as if this frame is never captured. This mode can be used in the scenario where the application doesn't want a 3A manual control capture to affect the subsequent auto 3A capture results.

Possible values:

Available values for this device:
android.control.availableModes

This key is available on all devices.

CONTROL_POST_RAW_SENSITIVITY_BOOST

Added in API level 24
public static final Key<Integer> CONTROL_POST_RAW_SENSITIVITY_BOOST

The amount of additional sensitivity boost applied to output images after RAW sensor data is captured.

Some camera devices support additional digital sensitivity boosting in the camera processing pipeline after sensor RAW image is captured. Such a boost will be applied to YUV/JPEG format output images but will not have effect on RAW output formats like RAW_SENSOR, RAW10, RAW12 or RAW_OPAQUE.

This key will be null for devices that do not support any RAW format outputs. For devices that do support RAW format outputs, this key will always present, and if a device does not support post RAW sensitivity boost, it will list 100 in this key.

If the camera device cannot apply the exact boost requested, it will reduce the boost to the nearest supported value. The final boost value used will be available in the output capture result.

For devices that support post RAW sensitivity boost, the YUV/JPEG output images of such device will have the total sensitivity of android.sensor.sensitivity * android.control.postRawSensitivityBoost / 100 The sensitivity of RAW format images will always be android.sensor.sensitivity

This control is only effective if android.control.aeMode or android.control.mode is set to OFF; otherwise the auto-exposure algorithm will override this value.

Units: ISO arithmetic units, the same as android.sensor.sensitivity

Range of valid values:
android.control.postRawSensitivityBoostRange

Optional - The value for this key may be null on some devices.

CONTROL_SCENE_MODE

Added in API level 21
public static final Key<Integer> CONTROL_SCENE_MODE

Control for which scene mode is currently active.

Scene modes are custom camera modes optimized for a certain set of conditions and capture settings.

This is the mode that that is active when android.control.mode == USE_SCENE_MODE. Aside from FACE_PRIORITY, these modes will disable android.control.aeMode, android.control.awbMode, and android.control.afMode while in use.

The interpretation and implementation of these scene modes is left to the implementor of the camera device. Their behavior will not be consistent across all devices, and any given device may only implement a subset of these modes.

Possible values:

Available values for this device:
android.control.availableSceneModes

This key is available on all devices.

See also:

CONTROL_SETTINGS_OVERRIDE

Added in API level 34
public static final Key<Integer> CONTROL_SETTINGS_OVERRIDE

The desired CaptureRequest settings override with which certain keys are applied earlier so that they can take effect sooner.

There are some CaptureRequest keys which can be applied earlier than others when controls within a CaptureRequest aren't required to take effect at the same time. One such example is zoom. Zoom can be applied at a later stage of the camera pipeline. As soon as the camera device receives the CaptureRequest, it can apply the requested zoom value onto an earlier request that's already in the pipeline, thus improves zoom latency.

This key's value in the capture result reflects whether the controls for this capture are overridden "by" a newer request. This means that if a capture request turns on settings override, the capture result of an earlier request will contain the key value of ZOOM. On the other hand, if a capture request has settings override turned on, but all newer requests have it turned off, the key's value in the capture result will be OFF because this capture isn't overridden by a newer capture. In the two examples below, the capture results columns illustrate the settingsOverride values in different scenarios.

Assuming the zoom settings override can speed up by 1 frame, below example illustrates the speed-up at the start of capture session:

Camera session created
 Request 1 (zoom=1.0x, override=ZOOM) ->
 Request 2 (zoom=1.2x, override=ZOOM) ->
 Request 3 (zoom=1.4x, override=ZOOM) ->  Result 1 (zoom=1.2x, override=ZOOM)
 Request 4 (zoom=1.6x, override=ZOOM) ->  Result 2 (zoom=1.4x, override=ZOOM)
 Request 5 (zoom=1.8x, override=ZOOM) ->  Result 3 (zoom=1.6x, override=ZOOM)
                                      ->  Result 4 (zoom=1.8x, override=ZOOM)
                                      ->  Result 5 (zoom=1.8x, override=OFF)
 

The application can turn on settings override and use zoom as normal. The example shows that the later zoom values (1.2x, 1.4x, 1.6x, and 1.8x) overwrite the zoom values (1.0x, 1.2x, 1.4x, and 1.8x) of earlier requests (#1, #2, #3, and #4).

The application must make sure the settings override doesn't interfere with user journeys requiring simultaneous application of all controls in CaptureRequest on the requested output targets. For example, if the application takes a still capture using CameraCaptureSession#capture, and the repeating request immediately sets a different zoom value using override, the inflight still capture could have its zoom value overwritten unexpectedly.

So the application is strongly recommended to turn off settingsOverride when taking still/burst captures, and turn it back on when there is only repeating viewfinder request and no inflight still/burst captures.

Below is the example demonstrating the transitions in and out of the settings override:

Request 1 (zoom=1.0x, override=OFF)
 Request 2 (zoom=1.2x, override=OFF)
 Request 3 (zoom=1.4x, override=ZOOM)  -> Result 1 (zoom=1.0x, override=OFF)
 Request 4 (zoom=1.6x, override=ZOOM)  -> Result 2 (zoom=1.4x, override=ZOOM)
 Request 5 (zoom=1.8x, override=OFF)   -> Result 3 (zoom=1.6x, override=ZOOM)
                                       -> Result 4 (zoom=1.6x, override=OFF)
                                       -> Result 5 (zoom=1.8x, override=OFF)
 

This example shows that:

  • The application "ramps in" settings override by setting the control to ZOOM. In the example, request #3 enables zoom settings override. Because the camera device can speed up applying zoom by 1 frame, the outputs of request #2 has 1.4x zoom, the value specified in request #3.
  • The application "ramps out" of settings override by setting the control to OFF. In the example, request #5 changes the override to OFF. Because request #4's zoom takes effect in result #3, result #4's zoom remains the same until new value takes effect in result #5.

Possible values:

Available values for this device:
android.control.availableSettingsOverrides

Optional - The value for this key may be null on some devices.

CONTROL_VIDEO_STABILIZATION_MODE

Added in API level 21
public static final Key<Integer> CONTROL_VIDEO_STABILIZATION_MODE

Whether video stabilization is active.

Video stabilization automatically warps images from the camera in order to stabilize motion between consecutive frames.

If enabled, video stabilization can modify the android.scaler.cropRegion to keep the video stream stabilized.

Switching between different video stabilization modes may take several frames to initialize, the camera device will report the current mode in capture result metadata. For example, When "ON" mode is requested, the video stabilization modes in the first several capture results may still be "OFF", and it will become "ON" when the initialization is done.

In addition, not all recording sizes or frame rates may be supported for stabilization by a device that reports stabilization support. It is guaranteed that an output targeting a MediaRecorder or MediaCodec will be stabilized if the recording resolution is less than or equal to 1920 x 1080 (width less than or equal to 1920, height less than or equal to 1080), and the recording frame rate is less than or equal to 30fps. At other sizes, the CaptureResult android.control.videoStabilizationMode field will return OFF if the recording output is not stabilized, or if there are no output Surface types that can be stabilized.

The application is strongly recommended to call SessionConfiguration.setSessionParameters(CaptureRequest) with the desired video stabilization mode before creating the capture session. Video stabilization mode is a session parameter on many devices. Specifying it at session creation time helps avoid reconfiguration delay caused by difference between the default value and the first CaptureRequest.

If a camera device supports both this mode and OIS (android.lens.opticalStabilizationMode), turning both modes on may produce undesirable interaction, so it is recommended not to enable both at the same time.

If video stabilization is set to "PREVIEW_STABILIZATION", android.lens.opticalStabilizationMode is overridden. The camera sub-system may choose to turn on hardware based image stabilization in addition to software based stabilization if it deems that appropriate. This key may be a part of the available session keys, which camera clients may query via CameraCharacteristics.getAvailableSessionKeys(). If this is the case, changing this key over the life-time of a capture session may cause delays / glitches.

Possible values:

This key is available on all devices.

CONTROL_ZOOM_RATIO

Added in API level 30
public static final Key<Float> CONTROL_ZOOM_RATIO

The desired zoom ratio

Instead of using android.scaler.cropRegion for zoom, the application can now choose to use this tag to specify the desired zoom level.

By using this control, the application gains a simpler way to control zoom, which can be a combination of optical and digital zoom. For example, a multi-camera system may contain more than one lens with different focal lengths, and the user can use optical zoom by switching between lenses. Using zoomRatio has benefits in the scenarios below:

  • Zooming in from a wide-angle lens to a telephoto lens: A floating-point ratio provides better precision compared to an integer value of android.scaler.cropRegion.
  • Zooming out from a wide lens to an ultrawide lens: zoomRatio supports zoom-out whereas android.scaler.cropRegion doesn't.

To illustrate, here are several scenarios of different zoom ratios, crop regions, and output streams, for a hypothetical camera device with an active array of size (2000,1500).

  • Camera Configuration:
    • Active array size: 2000x1500 (3 MP, 4:3 aspect ratio)
    • Output stream #1: 640x480 (VGA, 4:3 aspect ratio)
    • Output stream #2: 1280x720 (720p, 16:9 aspect ratio)
  • Case #1: 4:3 crop region with 2.0x zoom ratio
    • Zoomed field of view: 1/4 of original field of view
    • Crop region: Rect(0, 0, 2000, 1500) // (left, top, right, bottom) (post zoom)
  • 4:3 aspect ratio crop diagram
    • 640x480 stream source area: (0, 0, 2000, 1500) (equal to crop region)
    • 1280x720 stream source area: (0, 187, 2000, 1312) (letterboxed)
  • Case #2: 16:9 crop region with 2.0x zoom.
    • Zoomed field of view: 1/4 of original field of view
    • Crop region: Rect(0, 187, 2000, 1312)
    • 16:9 aspect ratio crop diagram
    • 640x480 stream source area: (250, 187, 1750, 1312) (pillarboxed)
    • 1280x720 stream source area: (0, 187, 2000, 1312) (equal to crop region)
  • Case #3: 1:1 crop region with 0.5x zoom out to ultrawide lens.
    • Zoomed field of view: 4x of original field of view (switched from wide lens to ultrawide lens)
    • Crop region: Rect(250, 0, 1750, 1500)
    • 1:1 aspect ratio crop diagram
    • 640x480 stream source area: (250, 187, 1750, 1312) (letterboxed)
    • 1280x720 stream source area: (250, 328, 1750, 1172) (letterboxed)

As seen from the graphs above, the coordinate system of cropRegion now changes to the effective after-zoom field-of-view, and is represented by the rectangle of (0, 0, activeArrayWith, activeArrayHeight). The same applies to AE/AWB/AF regions, and faces. This coordinate system change isn't applicable to RAW capture and its related metadata such as intrinsicCalibration and lensShadingMap.

Using the same hypothetical example above, and assuming output stream #1 (640x480) is the viewfinder stream, the application can achieve 2.0x zoom in one of two ways:

  • zoomRatio = 2.0, scaler.cropRegion = (0, 0, 2000, 1500)
  • zoomRatio = 1.0 (default), scaler.cropRegion = (500, 375, 1500, 1125)

If the application intends to set aeRegions to be top-left quarter of the viewfinder field-of-view, the android.control.aeRegions should be set to (0, 0, 1000, 750) with zoomRatio set to 2.0. Alternatively, the application can set aeRegions to the equivalent region of (500, 375, 1000, 750) for zoomRatio of 1.0. If the application doesn't explicitly set android.control.zoomRatio, its value defaults to 1.0.

One limitation of controlling zoom using zoomRatio is that the android.scaler.cropRegion must only be used for letterboxing or pillarboxing of the sensor active array, and no FREEFORM cropping can be used with android.control.zoomRatio other than 1.0. If android.control.zoomRatio is not 1.0, and android.scaler.cropRegion is set to be windowboxing, the camera framework will override the android.scaler.cropRegion to be the active array.

In the capture request, if the application sets android.control.zoomRatio to a value != 1.0, the android.control.zoomRatio tag in the capture result reflects the effective zoom ratio achieved by the camera device, and the android.scaler.cropRegion adjusts for additional crops that are not zoom related. Otherwise, if the application sets android.control.zoomRatio to 1.0, or does not set it at all, the android.control.zoomRatio tag in the result metadata will also be 1.0.

When the application requests a physical stream for a logical multi-camera, the android.control.zoomRatio in the physical camera result metadata will be 1.0, and the android.scaler.cropRegion tag reflects the amount of zoom and crop done by the physical camera device.

Range of valid values:
android.control.zoomRatioRange

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

DISTORTION_CORRECTION_MODE

Added in API level 28
public static final Key<Integer> DISTORTION_CORRECTION_MODE

Mode of operation for the lens distortion correction block.

The lens distortion correction block attempts to improve image quality by fixing radial, tangential, or other geometric aberrations in the camera device's optics. If available, the android.lens.distortion field documents the lens's distortion parameters.

OFF means no distortion correction is done.

FAST/HIGH_QUALITY both mean camera device determined distortion correction will be applied. HIGH_QUALITY mode indicates that the camera device will use the highest-quality correction algorithms, even if it slows down capture rate. FAST means the camera device will not slow down capture rate when applying correction. FAST may be the same as OFF if any correction at all would slow down capture rate. Every output stream will have a similar amount of enhancement applied.

The correction only applies to processed outputs such as YUV, Y8, JPEG, or DEPTH16; it is not applied to any RAW output.

This control will be on by default on devices that support this control. Applications disabling distortion correction need to pay extra attention with the coordinate system of metering regions, crop region, and face rectangles. When distortion correction is OFF, metadata coordinates follow the coordinate system of android.sensor.info.preCorrectionActiveArraySize. When distortion is not OFF, metadata coordinates follow the coordinate system of android.sensor.info.activeArraySize. The camera device will map these metadata fields to match the corrected image produced by the camera device, for both capture requests and results. However, this mapping is not very precise, since rectangles do not generally map to rectangles when corrected. Only linear scaling between the active array and precorrection active array coordinates is performed. Applications that require precise correction of metadata need to undo that linear scaling, and apply a more complete correction that takes into the account the app's own requirements.

The full list of metadata that is affected in this way by distortion correction is:

Possible values:

Available values for this device:
android.distortionCorrection.availableModes

Optional - The value for this key may be null on some devices.

EDGE_MODE

Added in API level 21
public static final Key<Integer> EDGE_MODE

Operation mode for edge enhancement.

Edge enhancement improves sharpness and details in the captured image. OFF means no enhancement will be applied by the camera device.

FAST/HIGH_QUALITY both mean camera device determined enhancement will be applied. HIGH_QUALITY mode indicates that the camera device will use the highest-quality enhancement algorithms, even if it slows down capture rate. FAST means the camera device will not slow down capture rate when applying edge enhancement. FAST may be the same as OFF if edge enhancement will slow down capture rate. Every output stream will have a similar amount of enhancement applied.

ZERO_SHUTTER_LAG is meant to be used by applications that maintain a continuous circular buffer of high-resolution images during preview and reprocess image(s) from that buffer into a final capture when triggered by the user. In this mode, the camera device applies edge enhancement to low-resolution streams (below maximum recording resolution) to maximize preview quality, but does not apply edge enhancement to high-resolution streams, since those will be reprocessed later if necessary.

For YUV_REPROCESSING, these FAST/HIGH_QUALITY modes both mean that the camera device will apply FAST/HIGH_QUALITY YUV-domain edge enhancement, respectively. The camera device may adjust its internal edge enhancement parameters for best image quality based on the android.reprocess.effectiveExposureFactor, if it is set.

Possible values:

Available values for this device:
android.edge.availableEdgeModes

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

EXTENSION_CURRENT_TYPE

Added in API level 34
public static final Key<Integer> EXTENSION_CURRENT_TYPE

Contains the extension type of the currently active extension

The capture result will only be supported and included by camera extension sessions. In case the extension session was configured to use AUTO, then the extension type value will indicate the currently active extension like HDR, NIGHT etc. , and will never return AUTO. In case the extension session was configured to use an extension different from AUTO, then the result type will always match with the configured extension type.

Range of valid values:
Extension type value listed in CameraExtensionCharacteristics

Optional - The value for this key may be null on some devices.

EXTENSION_STRENGTH

Added in API level 34
public static final Key<Integer> EXTENSION_STRENGTH

Strength of the extension post-processing effect

This control allows Camera extension clients to configure the strength of the applied extension effect. Strength equal to 0 means that the extension must not apply any post-processing and return a regular captured frame. Strength equal to 100 is the maximum level of post-processing. Values between 0 and 100 will have different effect depending on the extension type as described below:

  • BOKEH - the strength is expected to control the amount of blur.
  • HDR and NIGHT - the strength can control the amount of images fused and the brightness of the final image.
  • FACE_RETOUCH - the strength value will control the amount of cosmetic enhancement and skin smoothing.

The control will be supported if the capture request key is part of the list generated by CameraExtensionCharacteristics.getAvailableCaptureRequestKeys(int). The control is only defined and available to clients sending capture requests via CameraExtensionSession. If the client doesn't specify the extension strength value, then a default value will be set by the extension. Clients can retrieve the default value by checking the corresponding capture result.

Range of valid values:
0 - 100

Optional - The value for this key may be null on some devices.

FLASH_MODE

Added in API level 21
public static final Key<Integer> FLASH_MODE

The desired mode for for the camera device's flash control.

This control is only effective when flash unit is available (android.flash.info.available == true).

When this control is used, the android.control.aeMode must be set to ON or OFF. Otherwise, the camera device auto-exposure related flash control (ON_AUTO_FLASH, ON_ALWAYS_FLASH, or ON_AUTO_FLASH_REDEYE) will override this control.

When set to OFF, the camera device will not fire flash for this capture.

When set to SINGLE, the camera device will fire flash regardless of the camera device's auto-exposure routine's result. When used in still capture case, this control should be used along with auto-exposure (AE) precapture metering sequence (android.control.aePrecaptureTrigger), otherwise, the image may be incorrectly exposed.

When set to TORCH, the flash will be on continuously. This mode can be used for use cases such as preview, auto-focus assist, still capture, or video recording.

The flash status will be reported by android.flash.state in the capture result metadata.

Possible values:

This key is available on all devices.

FLASH_STATE

Added in API level 21
public static final Key<Integer> FLASH_STATE

Current state of the flash unit.

When the camera device doesn't have flash unit (i.e. android.flash.info.available == false), this state will always be UNAVAILABLE. Other states indicate the current flash status.

In certain conditions, this will be available on LEGACY devices:

In all other conditions the state will not be available on LEGACY devices (i.e. it will be null).

Possible values:

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

FLASH_STRENGTH_LEVEL

Added in API level 35
public static final Key<Integer> FLASH_STRENGTH_LEVEL

Flash strength level to be used when manual flash control is active.

Flash strength level to use in capture mode i.e. when the applications control flash with either SINGLE or TORCH mode.

Use android.flash.singleStrengthMaxLevel and android.flash.torchStrengthMaxLevel to check whether the device supports flash strength control or not. If the values of android.flash.singleStrengthMaxLevel and android.flash.torchStrengthMaxLevel are greater than 1, then the device supports manual flash strength control.

If the android.flash.mode == TORCH the value must be >= 1 and <= android.flash.torchStrengthMaxLevel. If the application doesn't set the key and android.flash.torchStrengthMaxLevel > 1, then the flash will be fired at the default level set by HAL in android.flash.torchStrengthDefaultLevel. If the android.flash.mode == SINGLE, then the value must be >= 1 and <= android.flash.singleStrengthMaxLevel. If the application does not set this key and android.flash.singleStrengthMaxLevel > 1, then the flash will be fired at the default level set by HAL in android.flash.singleStrengthDefaultLevel. If android.control.aeMode is set to any of ON_AUTO_FLASH, ON_ALWAYS_FLASH, ON_AUTO_FLASH_REDEYE, ON_EXTERNAL_FLASH values, then the strengthLevel will be ignored.

Range of valid values:
[1-android.flash.torchStrengthMaxLevel] when the android.flash.mode is set to TORCH; [1-android.flash.singleStrengthMaxLevel] when the android.flash.mode is set to SINGLE

This key is available on all devices.

HOT_PIXEL_MODE

Added in API level 21
public static final Key<Integer> HOT_PIXEL_MODE

Operational mode for hot pixel correction.

Hotpixel correction interpolates out, or otherwise removes, pixels that do not accurately measure the incoming light (i.e. pixels that are stuck at an arbitrary value or are oversensitive).

Possible values:

Available values for this device:
android.hotPixel.availableHotPixelModes

Optional - The value for this key may be null on some devices.

JPEG_GPS_LOCATION

Added in API level 21
public static final Key<Location> JPEG_GPS_LOCATION

A location object to use when generating image GPS metadata.

Setting a location object in a request will include the GPS coordinates of the location into any JPEG images captured based on the request. These coordinates can then be viewed by anyone who receives the JPEG image.

This tag is also used for HEIC image capture.

This key is available on all devices.

JPEG_ORIENTATION

Added in API level 21
public static final Key<Integer> JPEG_ORIENTATION

The orientation for a JPEG image.

The clockwise rotation angle in degrees, relative to the orientation to the camera, that the JPEG picture needs to be rotated by, to be viewed upright.

Camera devices may either encode this value into the JPEG EXIF header, or rotate the image data to match this orientation. When the image data is rotated, the thumbnail data will also be rotated. Additionally, in the case where the image data is rotated, Image.getWidth() and Image.getHeight() will not be updated to reflect the height and width of the rotated image.

Note that this orientation is relative to the orientation of the camera sensor, given by android.sensor.orientation.

To translate from the device orientation given by the Android sensor APIs for camera sensors which are not EXTERNAL, the following sample code may be used:

private int getJpegOrientation(CameraCharacteristics c, int deviceOrientation) {
     if (deviceOrientation == android.view.OrientationEventListener.ORIENTATION_UNKNOWN) return 0;
     int sensorOrientation = c.get(CameraCharacteristics.SENSOR_ORIENTATION);

     // Round device orientation to a multiple of 90
     deviceOrientation = (deviceOrientation + 45) / 90 * 90;

     // Reverse device orientation for front-facing cameras
     boolean facingFront = c.get(CameraCharacteristics.LENS_FACING) == CameraCharacteristics.LENS_FACING_FRONT;
     if (facingFront) deviceOrientation = -deviceOrientation;

     // Calculate desired JPEG orientation relative to camera orientation to make
     // the image upright relative to the device orientation
     int jpegOrientation = (sensorOrientation + deviceOrientation + 360) % 360;

     return jpegOrientation;
 }
 

For EXTERNAL cameras the sensor orientation will always be set to 0 and the facing will also be set to EXTERNAL. The above code is not relevant in such case.

This tag is also used to describe the orientation of the HEIC image capture, in which case the rotation is reflected by EXIF orientation flag, and not by rotating the image data itself.

Units: Degrees in multiples of 90

Range of valid values:
0, 90, 180, 270

This key is available on all devices.

JPEG_QUALITY

Added in API level 21
public static final Key<Byte> JPEG_QUALITY

Compression quality of the final JPEG image.

85-95 is typical usage range. This tag is also used to describe the quality of the HEIC image capture.

Range of valid values:
1-100; larger is higher quality

This key is available on all devices.

JPEG_THUMBNAIL_QUALITY

Added in API level 21
public static final Key<Byte> JPEG_THUMBNAIL_QUALITY

Compression quality of JPEG thumbnail.

This tag is also used to describe the quality of the HEIC image capture.

Range of valid values:
1-100; larger is higher quality

This key is available on all devices.

JPEG_THUMBNAIL_SIZE

Added in API level 21
public static final Key<Size> JPEG_THUMBNAIL_SIZE

Resolution of embedded JPEG thumbnail.

When set to (0, 0) value, the JPEG EXIF will not contain thumbnail, but the captured JPEG will still be a valid image.

For best results, when issuing a request for a JPEG image, the thumbnail size selected should have the same aspect ratio as the main JPEG output.

If the thumbnail image aspect ratio differs from the JPEG primary image aspect ratio, the camera device creates the thumbnail by cropping it from the primary image. For example, if the primary image has 4:3 aspect ratio, the thumbnail image has 16:9 aspect ratio, the primary image will be cropped vertically (letterbox) to generate the thumbnail image. The thumbnail image will always have a smaller Field Of View (FOV) than the primary image when aspect ratios differ.

When an android.jpeg.orientation of non-zero degree is requested, the camera device will handle thumbnail rotation in one of the following ways:

  • Set the EXIF orientation flag and keep jpeg and thumbnail image data unrotated.
  • Rotate the jpeg and thumbnail image data and not set EXIF orientation flag. In this case, LIMITED or FULL hardware level devices will report rotated thumbnail size in capture result, so the width and height will be interchanged if 90 or 270 degree orientation is requested. LEGACY device will always report unrotated thumbnail size.

The tag is also used as thumbnail size for HEIC image format capture, in which case the the thumbnail rotation is reflected by EXIF orientation flag, and not by rotating the thumbnail data itself.

Range of valid values:
android.jpeg.availableThumbnailSizes

This key is available on all devices.

LENS_APERTURE

Added in API level 21
public static final Key<Float> LENS_APERTURE

The desired lens aperture size, as a ratio of lens focal length to the effective aperture diameter.

Setting this value is only supported on the camera devices that have a variable aperture lens.

When this is supported and android.control.aeMode is OFF, this can be set along with android.sensor.exposureTime, android.sensor.sensitivity, and android.sensor.frameDuration to achieve manual exposure control.

The requested aperture value may take several frames to reach the requested value; the camera device will report the current (intermediate) aperture size in capture result metadata while the aperture is changing. While the aperture is still changing, android.lens.state will be set to MOVING.

When this is supported and android.control.aeMode is one of the ON modes, this will be overridden by the camera device auto-exposure algorithm, the overridden values are then provided back to the user in the corresponding result.

Units: The f-number (f/N)

Range of valid values:
android.lens.info.availableApertures

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

LENS_DISTORTION

Added in API level 28
public static final Key<float[]> LENS_DISTORTION

The correction coefficients to correct for this camera device's radial and tangential lens distortion.

Replaces the deprecated android.lens.radialDistortion field, which was inconsistently defined.

Three radial distortion coefficients [kappa_1, kappa_2, kappa_3] and two tangential distortion coefficients [kappa_4, kappa_5] that can be used to correct the lens's geometric distortion with the mapping equations:

 x_c = x_i * ( 1 + kappa_1 * r^2 + kappa_2 * r^4 + kappa_3 * r^6 ) +
        kappa_4 * (2 * x_i * y_i) + kappa_5 * ( r^2 + 2 * x_i^2 )
  y_c = y_i * ( 1 + kappa_1 * r^2 + kappa_2 * r^4 + kappa_3 * r^6 ) +
        kappa_5 * (2 * x_i * y_i) + kappa_4 * ( r^2 + 2 * y_i^2 )
 

Here, [x_c, y_c] are the coordinates to sample in the input image that correspond to the pixel values in the corrected image at the coordinate [x_i, y_i]:

 correctedImage(x_i, y_i) = sample_at(x_c, y_c, inputImage)
 

The pixel coordinates are defined in a coordinate system related to the android.lens.intrinsicCalibration calibration fields; see that entry for details of the mapping stages. Both [x_i, y_i] and [x_c, y_c] have (0,0) at the lens optical center [c_x, c_y], and the range of the coordinates depends on the focal length terms of the intrinsic calibration.

Finally, r represents the radial distance from the optical center, r^2 = x_i^2 + y_i^2.

The distortion model used is the Brown-Conrady model.

Units: Unitless coefficients.

Optional - The value for this key may be null on some devices.

Permission Manifest.permission.CAMERA is needed to access this property

LENS_FILTER_DENSITY

Added in API level 21
public static final Key<Float> LENS_FILTER_DENSITY

The desired setting for the lens neutral density filter(s).

This control will not be supported on most camera devices.

Lens filters are typically used to lower the amount of light the sensor is exposed to (measured in steps of EV). As used here, an EV step is the standard logarithmic representation, which are non-negative, and inversely proportional to the amount of light hitting the sensor. For example, setting this to 0 would result in no reduction of the incoming light, and setting this to 2 would mean that the filter is set to reduce incoming light by two stops (allowing 1/4 of the prior amount of light to the sensor).

It may take several frames before the lens filter density changes to the requested value. While the filter density is still changing, android.lens.state will be set to MOVING.

Units: Exposure Value (EV)

Range of valid values:
android.lens.info.availableFilterDensities

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

LENS_FOCAL_LENGTH

Added in API level 21
public static final Key<Float> LENS_FOCAL_LENGTH

The desired lens focal length; used for optical zoom.

This setting controls the physical focal length of the camera device's lens. Changing the focal length changes the field of view of the camera device, and is usually used for optical zoom.

Like android.lens.focusDistance and android.lens.aperture, this setting won't be applied instantaneously, and it may take several frames before the lens can change to the requested focal length. While the focal length is still changing, android.lens.state will be set to MOVING.

Optical zoom via this control will not be supported on most devices. Starting from API level 30, the camera device may combine optical and digital zoom through the android.control.zoomRatio control.

Units: Millimeters

Range of valid values:
android.lens.info.availableFocalLengths

This key is available on all devices.

LENS_FOCUS_DISTANCE

Added in API level 21
public static final Key<Float> LENS_FOCUS_DISTANCE

Desired distance to plane of sharpest focus, measured from frontmost surface of the lens.

Should be zero for fixed-focus cameras

Units: See android.lens.info.focusDistanceCalibration for details

Range of valid values:
>= 0

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

LENS_FOCUS_RANGE

Added in API level 21
public static final Key<Pair<FloatFloat>> LENS_FOCUS_RANGE

The range of scene distances that are in sharp focus (depth of field).

If variable focus not supported, can still report fixed depth of field range

Units: A pair of focus distances in diopters: (near, far); see android.lens.info.focusDistanceCalibration for details.

Range of valid values:
>=0

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

LENS_INTRINSIC_CALIBRATION

Added in API level 23
public static final Key<float[]> LENS_INTRINSIC_CALIBRATION

The parameters for this camera device's intrinsic calibration.

The five calibration parameters that describe the transform from camera-centric 3D coordinates to sensor pixel coordinates:

[f_x, f_y, c_x, c_y, s]
 

Where f_x and f_y are the horizontal and vertical focal lengths, [c_x, c_y] is the position of the optical axis, and s is a skew parameter for the sensor plane not being aligned with the lens plane.

These are typically used within a transformation matrix K:

K = [ f_x,   s, c_x,
        0, f_y, c_y,
        0    0,   1 ]
 

which can then be combined with the camera pose rotation R and translation t (android.lens.poseRotation and android.lens.poseTranslation, respectively) to calculate the complete transform from world coordinates to pixel coordinates:

P = [ K 0   * [ R -Rt
      0 1 ]      0 1 ]
 

(Note the negation of poseTranslation when mapping from camera to world coordinates, and multiplication by the rotation).

With p_w being a point in the world coordinate system and p_s being a point in the camera active pixel array coordinate system, and with the mapping including the homogeneous division by z:

 p_h = (x_h, y_h, z_h) = P p_w
 p_s = p_h / z_h
 

so [x_s, y_s] is the pixel coordinates of the world point, z_s = 1, and w_s is a measurement of disparity (depth) in pixel coordinates.

Note that the coordinate system for this transform is the android.sensor.info.preCorrectionActiveArraySize system, where (0,0) is the top-left of the preCorrectionActiveArraySize rectangle. Once the pose and intrinsic calibration transforms have been applied to a world point, then the android.lens.distortion transform needs to be applied, and the result adjusted to be in the android.sensor.info.activeArraySize coordinate system (where (0, 0) is the top-left of the activeArraySize rectangle), to determine the final pixel coordinate of the world point for processed (non-RAW) output buffers.

For camera devices, the center of pixel (x,y) is located at coordinate (x + 0.5, y + 0.5). So on a device with a precorrection active array of size (10,10), the valid pixel indices go from (0,0)-(9,9), and an perfectly-built camera would have an optical center at the exact center of the pixel grid, at coordinates (5.0, 5.0), which is the top-left corner of pixel (5,5).

Units: Pixels in the android.sensor.info.preCorrectionActiveArraySize coordinate system.

Optional - The value for this key may be null on some devices.

Permission Manifest.permission.CAMERA is needed to access this property

LENS_OPTICAL_STABILIZATION_MODE

Added in API level 21
public static final Key<Integer> LENS_OPTICAL_STABILIZATION_MODE

Sets whether the camera device uses optical image stabilization (OIS) when capturing images.

OIS is used to compensate for motion blur due to small movements of the camera during capture. Unlike digital image stabilization (android.control.videoStabilizationMode), OIS makes use of mechanical elements to stabilize the camera sensor, and thus allows for longer exposure times before camera shake becomes apparent.

Switching between different optical stabilization modes may take several frames to initialize, the camera device will report the current mode in capture result metadata. For example, When "ON" mode is requested, the optical stabilization modes in the first several capture results may still be "OFF", and it will become "ON" when the initialization is done.

If a camera device supports both OIS and digital image stabilization (android.control.videoStabilizationMode), turning both modes on may produce undesirable interaction, so it is recommended not to enable both at the same time.

If android.control.videoStabilizationMode is set to "PREVIEW_STABILIZATION", android.lens.opticalStabilizationMode is overridden. The camera sub-system may choose to turn on hardware based image stabilization in addition to software based stabilization if it deems that appropriate. This key's value in the capture result will reflect which OIS mode was chosen.

Not all devices will support OIS; see android.lens.info.availableOpticalStabilization for available controls.

Possible values:

Available values for this device:
android.lens.info.availableOpticalStabilization

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

LENS_POSE_ROTATION

Added in API level 23
public static final Key<float[]> LENS_POSE_ROTATION

The orientation of the camera relative to the sensor coordinate system.

The four coefficients that describe the quaternion rotation from the Android sensor coordinate system to a camera-aligned coordinate system where the X-axis is aligned with the long side of the image sensor, the Y-axis is aligned with the short side of the image sensor, and the Z-axis is aligned with the optical axis of the sensor.

To convert from the quaternion coefficients (x,y,z,w) to the axis of rotation (a_x, a_y, a_z) and rotation amount theta, the following formulas can be used:

 theta = 2 * acos(w)
 a_x = x / sin(theta/2)
 a_y = y / sin(theta/2)
 a_z = z / sin(theta/2)
 

To create a 3x3 rotation matrix that applies the rotation defined by this quaternion, the following matrix can be used:

R = [ 1 - 2y^2 - 2z^2,       2xy - 2zw,       2xz + 2yw,
            2xy + 2zw, 1 - 2x^2 - 2z^2,       2yz - 2xw,
            2xz - 2yw,       2yz + 2xw, 1 - 2x^2 - 2y^2 ]
 

This matrix can then be used to apply the rotation to a column vector point with

p' = Rp

where p is in the device sensor coordinate system, and p' is in the camera-oriented coordinate system.

If android.lens.poseReference is UNDEFINED, the quaternion rotation cannot be accurately represented by the camera device, and will be represented by default values matching its default facing.

Units: Quaternion coefficients

Optional - The value for this key may be null on some devices.

Permission Manifest.permission.CAMERA is needed to access this property

LENS_POSE_TRANSLATION

Added in API level 23
public static final Key<float[]> LENS_POSE_TRANSLATION

Position of the camera optical center.

The position of the camera device's lens optical center, as a three-dimensional vector (x,y,z).

Prior to Android P, or when android.lens.poseReference is PRIMARY_CAMERA, this position is relative to the optical center of the largest camera device facing in the same direction as this camera, in the Android sensor coordinate axes. Note that only the axis definitions are shared with the sensor coordinate system, but not the origin.

If this device is the largest or only camera device with a given facing, then this position will be (0, 0, 0); a camera device with a lens optical center located 3 cm from the main sensor along the +X axis (to the right from the user's perspective) will report (0.03, 0, 0). Note that this means that, for many computer vision applications, the position needs to be negated to convert it to a translation from the camera to the origin.

To transform a pixel coordinates between two cameras facing the same direction, first the source camera android.lens.distortion must be corrected for. Then the source camera android.lens.intrinsicCalibration needs to be applied, followed by the android.lens.poseRotation of the source camera, the translation of the source camera relative to the destination camera, the android.lens.poseRotation of the destination camera, and finally the inverse of android.lens.intrinsicCalibration of the destination camera. This obtains a radial-distortion-free coordinate in the destination camera pixel coordinates.

To compare this against a real image from the destination camera, the destination camera image then needs to be corrected for radial distortion before comparison or sampling.

When android.lens.poseReference is GYROSCOPE, then this position is relative to the center of the primary gyroscope on the device. The axis definitions are the same as with PRIMARY_CAMERA.

When android.lens.poseReference is UNDEFINED, this position cannot be accurately represented by the camera device, and will be represented as (0, 0, 0).

When android.lens.poseReference is AUTOMOTIVE, then this position is relative to the origin of the automotive sensor coordinate system, which is at the center of the rear axle.

Units: Meters

Optional - The value for this key may be null on some devices.

Permission Manifest.permission.CAMERA is needed to access this property

LENS_RADIAL_DISTORTION

Added in API level 23
Deprecated in API level 28
public static final Key<float[]> LENS_RADIAL_DISTORTION

This field was deprecated in API level 28.

This field was inconsistently defined in terms of its normalization. Use android.lens.distortion instead.

The correction coefficients to correct for this camera device's radial and tangential lens distortion.

Four radial distortion coefficients [kappa_0, kappa_1, kappa_2, kappa_3] and two tangential distortion coefficients [kappa_4, kappa_5] that can be used to correct the lens's geometric distortion with the mapping equations:

 x_c = x_i * ( kappa_0 + kappa_1 * r^2 + kappa_2 * r^4 + kappa_3 * r^6 ) +
        kappa_4 * (2 * x_i * y_i) + kappa_5 * ( r^2 + 2 * x_i^2 )
  y_c = y_i * ( kappa_0 + kappa_1 * r^2 + kappa_2 * r^4 + kappa_3 * r^6 ) +
        kappa_5 * (2 * x_i * y_i) + kappa_4 * ( r^2 + 2 * y_i^2 )
 

Here, [x_c, y_c] are the coordinates to sample in the input image that correspond to the pixel values in the corrected image at the coordinate [x_i, y_i]:

 correctedImage(x_i, y_i) = sample_at(x_c, y_c, inputImage)
 

The pixel coordinates are defined in a normalized coordinate system related to the android.lens.intrinsicCalibration calibration fields. Both [x_i, y_i] and [x_c, y_c] have (0,0) at the lens optical center [c_x, c_y]. The maximum magnitudes of both x and y coordinates are normalized to be 1 at the edge further from the optical center, so the range for both dimensions is -1 <= x <= 1.

Finally, r represents the radial distance from the optical center, r^2 = x_i^2 + y_i^2, and its magnitude is therefore no larger than |r| <= sqrt(2).

The distortion model used is the Brown-Conrady model.

Units: Unitless coefficients.

Optional - The value for this key may be null on some devices.

Permission Manifest.permission.CAMERA is needed to access this property

LENS_STATE

Added in API level 21
public static final Key<Integer> LENS_STATE

Current lens status.

For lens parameters android.lens.focalLength, android.lens.focusDistance, android.lens.filterDensity and android.lens.aperture, when changes are requested, they may take several frames to reach the requested values. This state indicates the current status of the lens parameters.

When the state is STATIONARY, the lens parameters are not changing. This could be either because the parameters are all fixed, or because the lens has had enough time to reach the most recently-requested values. If all these lens parameters are not changeable for a camera device, as listed below:

Then this state will always be STATIONARY.

When the state is MOVING, it indicates that at least one of the lens parameters is changing.

Possible values:

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

LOGICAL_MULTI_CAMERA_ACTIVE_PHYSICAL_ID

Added in API level 29
public static final Key<String> LOGICAL_MULTI_CAMERA_ACTIVE_PHYSICAL_ID

String containing the ID of the underlying active physical camera.

The ID of the active physical camera that's backing the logical camera. All camera streams and metadata that are not physical camera specific will be originating from this physical camera.

For a logical camera made up of physical cameras where each camera's lenses have different characteristics, the camera device may choose to switch between the physical cameras when application changes FOCAL_LENGTH or SCALER_CROP_REGION. At the time of lens switch, this result metadata reflects the new active physical camera ID.

This key will be available if the camera device advertises this key via CameraCharacteristics.getAvailableCaptureResultKeys(). When available, this must be one of valid physical IDs backing this logical multi-camera. If this key is not available for a logical multi-camera, the camera device implementation may still switch between different active physical cameras based on use case, but the current active physical camera information won't be available to the application.

Optional - The value for this key may be null on some devices.

LOGICAL_MULTI_CAMERA_ACTIVE_PHYSICAL_SENSOR_CROP_REGION

Added in API level 35
public static final Key<Rect> LOGICAL_MULTI_CAMERA_ACTIVE_PHYSICAL_SENSOR_CROP_REGION

The current region of the active physical sensor that will be read out for this capture.

This capture result matches with android.scaler.cropRegion on non-logical single camera sensor devices. In case of logical cameras that can switch between several physical devices in response to android.control.zoomRatio, this capture result will not behave like android.scaler.cropRegion and android.control.zoomRatio, where the combination of both reflects the effective zoom and crop of the logical camera output. Instead, this capture result value will describe the zoom and crop of the active physical device. Some examples of when the value of this capture result will change include switches between different physical lenses, switches between regular and maximum resolution pixel mode and going through the device digital or optical range. This capture result is similar to android.scaler.cropRegion with respect to distortion correction. When the distortion correction mode is OFF, the coordinate system follows android.sensor.info.preCorrectionActiveArraySize, with (0, 0) being the top-left pixel of the pre-correction active array. When the distortion correction mode is not OFF, the coordinate system follows android.sensor.info.activeArraySize, with (0, 0) being the top-left pixel of the active array.

For camera devices with the CameraMetadata.REQUEST_AVAILABLE_CAPABILITIES_ULTRA_HIGH_RESOLUTION_SENSOR capability or devices where CameraCharacteristics#getAvailableCaptureRequestKeys lists CaptureRequest.SENSOR_PIXEL_MODE} , the current active physical device android.sensor.info.activeArraySizeMaximumResolution / android.sensor.info.preCorrectionActiveArraySizeMaximumResolution must be used as the coordinate system for requests where android.sensor.pixelMode is set to CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION.

Units: Pixel coordinates relative to android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize of the currently android.logicalMultiCamera.activePhysicalId depending on distortion correction capability and mode

Optional - The value for this key may be null on some devices.

NOISE_REDUCTION_MODE

Added in API level 21
public static final Key<Integer> NOISE_REDUCTION_MODE

Mode of operation for the noise reduction algorithm.

The noise reduction algorithm attempts to improve image quality by removing excessive noise added by the capture process, especially in dark conditions.

OFF means no noise reduction will be applied by the camera device, for both raw and YUV domain.

MINIMAL means that only sensor raw domain basic noise reduction is enabled ,to remove demosaicing or other processing artifacts. For YUV_REPROCESSING, MINIMAL is same as OFF. This mode is optional, may not be support by all devices. The application should check android.noiseReduction.availableNoiseReductionModes before using it.

FAST/HIGH_QUALITY both mean camera device determined noise filtering will be applied. HIGH_QUALITY mode indicates that the camera device will use the highest-quality noise filtering algorithms, even if it slows down capture rate. FAST means the camera device will not slow down capture rate when applying noise filtering. FAST may be the same as MINIMAL if MINIMAL is listed, or the same as OFF if any noise filtering will slow down capture rate. Every output stream will have a similar amount of enhancement applied.

ZERO_SHUTTER_LAG is meant to be used by applications that maintain a continuous circular buffer of high-resolution images during preview and reprocess image(s) from that buffer into a final capture when triggered by the user. In this mode, the camera device applies noise reduction to low-resolution streams (below maximum recording resolution) to maximize preview quality, but does not apply noise reduction to high-resolution streams, since those will be reprocessed later if necessary.

For YUV_REPROCESSING, these FAST/HIGH_QUALITY modes both mean that the camera device will apply FAST/HIGH_QUALITY YUV domain noise reduction, respectively. The camera device may adjust the noise reduction parameters for best image quality based on the android.reprocess.effectiveExposureFactor if it is set.

Possible values:

Available values for this device:
android.noiseReduction.availableNoiseReductionModes

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

REPROCESS_EFFECTIVE_EXPOSURE_FACTOR

Added in API level 23
public static final Key<Float> REPROCESS_EFFECTIVE_EXPOSURE_FACTOR

The amount of exposure time increase factor applied to the original output frame by the application processing before sending for reprocessing.

This is optional, and will be supported if the camera device supports YUV_REPROCESSING capability (android.request.availableCapabilities contains YUV_REPROCESSING).

For some YUV reprocessing use cases, the application may choose to filter the original output frames to effectively reduce the noise to the same level as a frame that was captured with longer exposure time. To be more specific, assuming the original captured images were captured with a sensitivity of S and an exposure time of T, the model in the camera device is that the amount of noise in the image would be approximately what would be expected if the original capture parameters had been a sensitivity of S/effectiveExposureFactor and an exposure time of T*effectiveExposureFactor, rather than S and T respectively. If the captured images were processed by the application before being sent for reprocessing, then the application may have used image processing algorithms and/or multi-frame image fusion to reduce the noise in the application-processed images (input images). By using the effectiveExposureFactor control, the application can communicate to the camera device the actual noise level improvement in the application-processed image. With this information, the camera device can select appropriate noise reduction and edge enhancement parameters to avoid excessive noise reduction (android.noiseReduction.mode) and insufficient edge enhancement (android.edge.mode) being applied to the reprocessed frames.

For example, for multi-frame image fusion use case, the application may fuse multiple output frames together to a final frame for reprocessing. When N image are fused into 1 image for reprocessing, the exposure time increase factor could be up to square root of N (based on a simple photon shot noise model). The camera device will adjust the reprocessing noise reduction and edge enhancement parameters accordingly to produce the best quality images.

This is relative factor, 1.0 indicates the application hasn't processed the input buffer in a way that affects its effective exposure time.

This control is only effective for YUV reprocessing capture request. For noise reduction reprocessing, it is only effective when android.noiseReduction.mode != OFF. Similarly, for edge enhancement reprocessing, it is only effective when android.edge.mode != OFF.

Units: Relative exposure time increase factor.

Range of valid values:
>= 1.0

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

REQUEST_PIPELINE_DEPTH

Added in API level 21
public static final Key<Byte> REQUEST_PIPELINE_DEPTH

Specifies the number of pipeline stages the frame went through from when it was exposed to when the final completed result was available to the framework.

Depending on what settings are used in the request, and what streams are configured, the data may undergo less processing, and some pipeline stages skipped.

See android.request.pipelineMaxDepth for more details.

Range of valid values:
<= android.request.pipelineMaxDepth

This key is available on all devices.

SCALER_CROP_REGION

Added in API level 21
public static final Key<Rect> SCALER_CROP_REGION

The desired region of the sensor to read out for this capture.

This control can be used to implement digital zoom.

For devices not supporting android.distortionCorrection.mode control, the coordinate system always follows that of android.sensor.info.activeArraySize, with (0, 0) being the top-left pixel of the active array.

For devices supporting android.distortionCorrection.mode control, the coordinate system depends on the mode being set. When the distortion correction mode is OFF, the coordinate system follows android.sensor.info.preCorrectionActiveArraySize, with (0, 0) being the top-left pixel of the pre-correction active array. When the distortion correction mode is not OFF, the coordinate system follows android.sensor.info.activeArraySize, with (0, 0) being the top-left pixel of the active array.

Output streams use this rectangle to produce their output, cropping to a smaller region if necessary to maintain the stream's aspect ratio, then scaling the sensor input to match the output's configured resolution.

The crop region is usually applied after the RAW to other color space (e.g. YUV) conversion. As a result RAW streams are not croppable unless supported by the camera device. See android.scaler.availableStreamUseCases#CROPPED_RAW for details.

For non-raw streams, any additional per-stream cropping will be done to maximize the final pixel area of the stream.

For example, if the crop region is set to a 4:3 aspect ratio, then 4:3 streams will use the exact crop region. 16:9 streams will further crop vertically (letterbox).

Conversely, if the crop region is set to a 16:9, then 4:3 outputs will crop horizontally (pillarbox), and 16:9 streams will match exactly. These additional crops will be centered within the crop region.

To illustrate, here are several scenarios of different crop regions and output streams, for a hypothetical camera device with an active array of size (2000,1500). Note that several of these examples use non-centered crop regions for ease of illustration; such regions are only supported on devices with FREEFORM capability (android.scaler.croppingType == FREEFORM), but this does not affect the way the crop rules work otherwise.

  • Camera Configuration:
    • Active array size: 2000x1500 (3 MP, 4:3 aspect ratio)
    • Output stream #1: 640x480 (VGA, 4:3 aspect ratio)
    • Output stream #2: 1280x720 (720p, 16:9 aspect ratio)
  • Case #1: 4:3 crop region with 2x digital zoom
    • Crop region: Rect(500, 375, 1500, 1125) // (left, top, right, bottom)
    • 4:3 aspect ratio crop diagram
    • 640x480 stream source area: (500, 375, 1500, 1125) (equal to crop region)
    • 1280x720 stream source area: (500, 469, 1500, 1031) (letterboxed)
  • Case #2: 16:9 crop region with ~1.5x digital zoom.
    • Crop region: Rect(500, 375, 1833, 1125)
    • 16:9 aspect ratio crop diagram
    • 640x480 stream source area: (666, 375, 1666, 1125) (pillarboxed)
    • 1280x720 stream source area: (500, 375, 1833, 1125) (equal to crop region)
  • Case #3: 1:1 crop region with ~2.6x digital zoom.
    • Crop region: Rect(500, 375, 1250, 1125)
    • 1:1 aspect ratio crop diagram
    • 640x480 stream source area: (500, 469, 1250, 1031) (letterboxed)
    • 1280x720 stream source area: (500, 543, 1250, 957) (letterboxed)
  • Case #4: Replace 640x480 stream with 1024x1024 stream, with 4:3 crop region:
    • Crop region: Rect(500, 375, 1500, 1125)
    • Square output, 4:3 aspect ratio crop diagram
    • 1024x1024 stream source area: (625, 375, 1375, 1125) (pillarboxed)
    • 1280x720 stream source area: (500, 469, 1500, 1031) (letterboxed)
    • Note that in this case, neither of the two outputs is a subset of the other, with each containing image data the other doesn't have.

If the coordinate system is android.sensor.info.activeArraySize, the width and height of the crop region cannot be set to be smaller than floor( activeArraySize.width / android.scaler.availableMaxDigitalZoom ) and floor( activeArraySize.height / android.scaler.availableMaxDigitalZoom ), respectively.

If the coordinate system is android.sensor.info.preCorrectionActiveArraySize, the width and height of the crop region cannot be set to be smaller than floor( preCorrectionActiveArraySize.width / android.scaler.availableMaxDigitalZoom ) and floor( preCorrectionActiveArraySize.height / android.scaler.availableMaxDigitalZoom ), respectively.

The camera device may adjust the crop region to account for rounding and other hardware requirements; the final crop region used will be included in the output capture result.

The camera sensor output aspect ratio depends on factors such as output stream combination and android.control.aeTargetFpsRange, and shouldn't be adjusted by using this control. And the camera device will treat different camera sensor output sizes (potentially with in-sensor crop) as the same crop of android.sensor.info.activeArraySize. As a result, the application shouldn't assume the maximum crop region always maps to the same aspect ratio or field of view for the sensor output.

Starting from API level 30, it's strongly recommended to use android.control.zoomRatio to take advantage of better support for zoom with logical multi-camera. The benefits include better precision with optical-digital zoom combination, and ability to do zoom-out from 1.0x. When using android.control.zoomRatio for zoom, the crop region in the capture request should be left as the default activeArray size. The coordinate system is post-zoom, meaning that the activeArraySize or preCorrectionActiveArraySize covers the camera device's field of view "after" zoom. See android.control.zoomRatio for details.

For camera devices with the CameraMetadata.REQUEST_AVAILABLE_CAPABILITIES_ULTRA_HIGH_RESOLUTION_SENSOR capability or devices where CameraCharacteristics#getAvailableCaptureRequestKeys lists CaptureRequest.SENSOR_PIXEL_MODE}

android.sensor.info.activeArraySizeMaximumResolution / android.sensor.info.preCorrectionActiveArraySizeMaximumResolution must be used as the coordinate system for requests where android.sensor.pixelMode is set to CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION.

Units: Pixel coordinates relative to android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

This key is available on all devices.

SCALER_RAW_CROP_REGION

Added in API level 34
public static final Key<Rect> SCALER_RAW_CROP_REGION

The region of the sensor that corresponds to the RAW read out for this capture when the stream use case of a RAW stream is set to CROPPED_RAW.

The coordinate system follows that of android.sensor.info.preCorrectionActiveArraySize.

This CaptureResult key will be set when the corresponding CaptureRequest has a RAW target with stream use case set to CameraMetadata.SCALER_AVAILABLE_STREAM_USE_CASES_CROPPED_RAW, otherwise it will be null. The value of this key specifies the region of the sensor used for the RAW capture and can be used to calculate the corresponding field of view of RAW streams. This field of view will always be >= field of view for (processed) non-RAW streams for the capture. Note: The region specified may not necessarily be centered.

For example: Assume a camera device has a pre correction active array size of {0, 0, 1500, 2000}. If the RAW_CROP_REGION is {500, 375, 1500, 1125}, that corresponds to a centered crop of 1/4th of the full field of view RAW stream.

The metadata keys which describe properties of RAW frames:

should be interpreted in the effective after raw crop field-of-view coordinate system. In this coordinate system, {android.sensor.info.preCorrectionActiveArraySize.left, android.sensor.info.preCorrectionActiveArraySize.top} corresponds to the the top left corner of the cropped RAW frame and {android.sensor.info.preCorrectionActiveArraySize.right, android.sensor.info.preCorrectionActiveArraySize.bottom} corresponds to the bottom right corner. Client applications must use the values of the keys in the CaptureResult metadata if present.

Crop regions android.scaler.cropRegion, AE/AWB/AF regions and face coordinates still use the android.sensor.info.activeArraySize coordinate system as usual.

Units: Pixel coordinates relative to android.sensor.info.activeArraySize or android.sensor.info.preCorrectionActiveArraySize depending on distortion correction capability and mode

Optional - The value for this key may be null on some devices.

SCALER_ROTATE_AND_CROP

Added in API level 31
public static final Key<Integer> SCALER_ROTATE_AND_CROP

Whether a rotation-and-crop operation is applied to processed outputs from the camera.

This control is primarily intended to help camera applications with no support for multi-window modes to work correctly on devices where multi-window scenarios are unavoidable, such as foldables or other devices with variable display geometry or more free-form window placement (such as laptops, which often place portrait-orientation apps in landscape with pillarboxing).

If supported, the default value is ROTATE_AND_CROP_AUTO, which allows the camera API to enable backwards-compatibility support for applications that do not support resizing / multi-window modes, when the device is in fact in a multi-window mode (such as inset portrait on laptops, or on a foldable device in some fold states). In addition, ROTATE_AND_CROP_NONE and ROTATE_AND_CROP_90 will always be available if this control is supported by the device. If not supported, devices API level 30 or higher will always list only ROTATE_AND_CROP_NONE.

When CROP_AUTO is in use, and the camera API activates backward-compatibility mode, several metadata fields will also be parsed differently to ensure that coordinates are correctly handled for features like drawing face detection boxes or passing in tap-to-focus coordinates. The camera API will convert positions in the active array coordinate system to/from the cropped-and-rotated coordinate system to make the operation transparent for applications. The following controls are affected:

Capture results will contain the actual value selected by the API; ROTATE_AND_CROP_AUTO will never be seen in a capture result.

Applications can also select their preferred cropping mode, either to opt out of the backwards-compatibility treatment, or to use the cropping feature themselves as needed. In this case, no coordinate translation will be done automatically, and all controls will continue to use the normal active array coordinates.

Cropping and rotating is done after the application of digital zoom (via either android.scaler.cropRegion or android.control.zoomRatio), but before each individual output is further cropped and scaled. It only affects processed outputs such as YUV, PRIVATE, and JPEG. It has no effect on RAW outputs.

When CROP_90 or CROP_270 are selected, there is a significant loss to the field of view. For example, with a 4:3 aspect ratio output of 1600x1200, CROP_90 will still produce 1600x1200 output, but these buffers are cropped from a vertical 3:4 slice at the center of the 4:3 area, then rotated to be 4:3, and then upscaled to 1600x1200. Only 56.25% of the original FOV is still visible. In general, for an aspect ratio of w:h, the crop and rotate operation leaves (h/w)^2 of the field of view visible. For 16:9, this is ~31.6%.

As a visual example, the figure below shows the effect of ROTATE_AND_CROP_90 on the outputs for the following parameters:

  • Sensor active array: 2000x1500
  • Crop region: top-left: (500, 375), size: (1000, 750) (4:3 aspect ratio)
  • Output streams: YUV 640x480 and YUV 1280x720
  • ROTATE_AND_CROP_90

Effect of ROTATE_AND_CROP_90

With these settings, the regions of the active array covered by the output streams are:

  • 640x480 stream crop: top-left: (219, 375), size: (562, 750)
  • 1280x720 stream crop: top-left: (289, 375), size: (422, 750)

Since the buffers are rotated, the buffers as seen by the application are:

  • 640x480 stream: top-left: (781, 375) on active array, size: (640, 480), downscaled 1.17x from sensor pixels
  • 1280x720 stream: top-left: (711, 375) on active array, size: (1280, 720), upscaled 1.71x from sensor pixels

Possible values:

Available values for this device:
android.scaler.availableRotateAndCropModes

Optional - The value for this key may be null on some devices.

SENSOR_DYNAMIC_BLACK_LEVEL

Added in API level 24
public static final Key<float[]> SENSOR_DYNAMIC_BLACK_LEVEL

A per-frame dynamic black level offset for each of the color filter arrangement (CFA) mosaic channels.

Camera sensor black levels may vary dramatically for different capture settings (e.g. android.sensor.sensitivity). The fixed black level reported by android.sensor.blackLevelPattern may be too inaccurate to represent the actual value on a per-frame basis. The camera device internal pipeline relies on reliable black level values to process the raw images appropriately. To get the best image quality, the camera device may choose to estimate the per frame black level values either based on optically shielded black regions (android.sensor.opticalBlackRegions) or its internal model.

This key reports the camera device estimated per-frame zero light value for each of the CFA mosaic channels in the camera sensor. The android.sensor.blackLevelPattern may only represent a coarse approximation of the actual black level values. This value is the black level used in camera device internal image processing pipeline and generally more accurate than the fixed black level values. However, since they are estimated values by the camera device, they may not be as accurate as the black level values calculated from the optical black pixels reported by android.sensor.opticalBlackRegions.

The values are given in the same order as channels listed for the CFA layout key (see android.sensor.info.colorFilterArrangement), i.e. the nth value given corresponds to the black level offset for the nth color channel listed in the CFA.

For a MONOCHROME camera, all of the 2x2 channels must have the same values.

This key will be available if android.sensor.opticalBlackRegions is available or the camera device advertises this key via CameraCharacteristics.getAvailableCaptureResultKeys().

Range of valid values:
>= 0 for each.

Optional - The value for this key may be null on some devices.

SENSOR_DYNAMIC_WHITE_LEVEL

Added in API level 24
public static final Key<Integer> SENSOR_DYNAMIC_WHITE_LEVEL

Maximum raw value output by sensor for this frame.

Since the android.sensor.blackLevelPattern may change for different capture settings (e.g., android.sensor.sensitivity), the white level will change accordingly. This key is similar to android.sensor.info.whiteLevel, but specifies the camera device estimated white level for each frame.

This key will be available if android.sensor.opticalBlackRegions is available or the camera device advertises this key via CameraCharacteristics.getAvailableCaptureRequestKeys().

Range of valid values:
>= 0

Optional - The value for this key may be null on some devices.

SENSOR_EXPOSURE_TIME

Added in API level 21
public static final Key<Long> SENSOR_EXPOSURE_TIME

Duration each pixel is exposed to light.

If the sensor can't expose this exact duration, it will shorten the duration exposed to the nearest possible value (rather than expose longer). The final exposure time used will be available in the output capture result.

This control is only effective if android.control.aeMode or android.control.mode is set to OFF; otherwise the auto-exposure algorithm will override this value.

Units: Nanoseconds

Range of valid values:
android.sensor.info.exposureTimeRange

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

SENSOR_FRAME_DURATION

Added in API level 21
public static final Key<Long> SENSOR_FRAME_DURATION

Duration from start of frame readout to start of next frame readout.

The maximum frame rate that can be supported by a camera subsystem is a function of many factors:

  • Requested resolutions of output image streams
  • Availability of binning / skipping modes on the imager
  • The bandwidth of the imager interface
  • The bandwidth of the various ISP processing blocks

Since these factors can vary greatly between different ISPs and sensors, the camera abstraction tries to represent the bandwidth restrictions with as simple a model as possible.

The model presented has the following characteristics:

  • The image sensor is always configured to output the smallest resolution possible given the application's requested output stream sizes. The smallest resolution is defined as being at least as large as the largest requested output stream size; the camera pipeline must never digitally upsample sensor data when the crop region covers the whole sensor. In general, this means that if only small output stream resolutions are configured, the sensor can provide a higher frame rate.
  • Since any request may use any or all the currently configured output streams, the sensor and ISP must be configured to support scaling a single capture to all the streams at the same time. This means the camera pipeline must be ready to produce the largest requested output size without any delay. Therefore, the overall frame rate of a given configured stream set is governed only by the largest requested stream resolution.
  • Using more than one output stream in a request does not affect the frame duration.
  • Certain format-streams may need to do additional background processing before data is consumed/produced by that stream. These processors can run concurrently to the rest of the camera pipeline, but cannot process more than 1 capture at a time.

The necessary information for the application, given the model above, is provided via StreamConfigurationMap.getOutputMinFrameDuration(int, Size). These are used to determine the maximum frame rate / minimum frame duration that is possible for a given stream configuration.

Specifically, the application can use the following rules to determine the minimum frame duration it can request from the camera device:

  1. Let the set of currently configured input/output streams be called S.
  2. Find the minimum frame durations for each stream in S, by looking it up in StreamConfigurationMap.getOutputMinFrameDuration(int, Size) (with its respective size/format). Let this set of frame durations be called F.
  3. For any given request R, the minimum frame duration allowed for R is the maximum out of all values in F. Let the streams used in R be called S_r.

If none of the streams in S_r have a stall time (listed in StreamConfigurationMap.getOutputStallDuration(int, Size) using its respective size/format), then the frame duration in F determines the steady state frame rate that the application will get if it uses R as a repeating request. Let this special kind of request be called Rsimple.

A repeating request Rsimple can be occasionally interleaved by a single capture of a new request Rstall (which has at least one in-use stream with a non-0 stall time) and if Rstall has the same minimum frame duration this will not cause a frame rate loss if all buffers from the previous Rstall have already been delivered.

For more details about stalling, see StreamConfigurationMap.getOutputStallDuration(int, Size).

This control is only effective if android.control.aeMode or android.control.mode is set to OFF; otherwise the auto-exposure algorithm will override this value.

Note: Prior to Android 13, this field was described as measuring the duration from start of frame exposure to start of next frame exposure, which doesn't reflect the definition from sensor manufacturer. A mobile sensor defines the frame duration as intervals between sensor readouts.

Units: Nanoseconds

Range of valid values:
See android.sensor.info.maxFrameDuration, StreamConfigurationMap. The duration is capped to max(duration, exposureTime + overhead).

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

SENSOR_GREEN_SPLIT

Added in API level 21
public static final Key<Float> SENSOR_GREEN_SPLIT

The worst-case divergence between Bayer green channels.

This value is an estimate of the worst case split between the Bayer green channels in the red and blue rows in the sensor color filter array.

The green split is calculated as follows:

  1. A 5x5 pixel (or larger) window W within the active sensor array is chosen. The term 'pixel' here is taken to mean a group of 4 Bayer mosaic channels (R, Gr, Gb, B). The location and size of the window chosen is implementation defined, and should be chosen to provide a green split estimate that is both representative of the entire image for this camera sensor, and can be calculated quickly.
  2. The arithmetic mean of the green channels from the red rows (mean_Gr) within W is computed.
  3. The arithmetic mean of the green channels from the blue rows (mean_Gb) within W is computed.
  4. The maximum ratio R of the two means is computed as follows: R = max((mean_Gr + 1)/(mean_Gb + 1), (mean_Gb + 1)/(mean_Gr + 1))

The ratio R is the green split divergence reported for this property, which represents how much the green channels differ in the mosaic pattern. This value is typically used to determine the treatment of the green mosaic channels when demosaicing.

The green split value can be roughly interpreted as follows:

  • R < 1.03 is a negligible split (<3% divergence).
  • 1.20 <= R >= 1.03 will require some software correction to avoid demosaic errors (3-20% divergence).
  • R > 1.20 will require strong software correction to produce a usable image (>20% divergence).

Starting from Android Q, this key will not be present for a MONOCHROME camera, even if the camera device has RAW capability.

Range of valid values:

>= 0

Optional - The value for this key may be null on some devices.

SENSOR_NEUTRAL_COLOR_POINT

Added in API level 21
public static final Key<Rational[]> SENSOR_NEUTRAL_COLOR_POINT

The estimated camera neutral color in the native sensor colorspace at the time of capture.

This value gives the neutral color point encoded as an RGB value in the native sensor color space. The neutral color point indicates the currently estimated white point of the scene illumination. It can be used to interpolate between the provided color transforms when processing raw sensor data.

The order of the values is R, G, B; where R is in the lowest index.

Starting from Android Q, this key will not be present for a MONOCHROME camera, even if the camera device has RAW capability.

Optional - The value for this key may be null on some devices.

SENSOR_NOISE_PROFILE

Added in API level 21
public static final Key<Pair[]<DoubleDouble>> SENSOR_NOISE_PROFILE

Noise model coefficients for each CFA mosaic channel.

This key contains two noise model coefficients for each CFA channel corresponding to the sensor amplification (S) and sensor readout noise (O). These are given as pairs of coefficients for each channel in the same order as channels listed for the CFA layout key (see android.sensor.info.colorFilterArrangement). This is represented as an array of Pair<Double, Double>, where the first member of the Pair at index n is the S coefficient and the second member is the O coefficient for the nth color channel in the CFA.

These coefficients are used in a two parameter noise model to describe the amount of noise present in the image for each CFA channel. The noise model used here is:

N(x) = sqrt(Sx + O)

Where x represents the recorded signal of a CFA channel normalized to the range [0, 1], and S and O are the noise model coefficients for that channel.

A more detailed description of the noise model can be found in the Adobe DNG specification for the NoiseProfile tag.

For a MONOCHROME camera, there is only one color channel. So the noise model coefficients will only contain one S and one O.

Optional - The value for this key may be null on some devices.

SENSOR_PIXEL_MODE

Added in API level 31
public static final Key<Integer> SENSOR_PIXEL_MODE

Switches sensor pixel mode between maximum resolution mode and default mode.

This key controls whether the camera sensor operates in CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION mode or not. By default, all camera devices operate in CameraMetadata.SENSOR_PIXEL_MODE_DEFAULT mode. When operating in CameraMetadata.SENSOR_PIXEL_MODE_DEFAULT mode, sensors would typically perform pixel binning in order to improve low light performance, noise reduction etc. However, in CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION mode, sensors typically operate in unbinned mode allowing for a larger image size. The stream configurations supported in CameraMetadata.SENSOR_PIXEL_MODE_MAXIMUM_RESOLUTION mode are also different from those of CameraMetadata.SENSOR_PIXEL_MODE_DEFAULT mode. They can be queried through CameraCharacteristics.get(Key) with CameraCharacteristics#SCALER_STREAM_CONFIGURATION_MAP_MAXIMUM_RESOLUTION. Unless reported by both StreamConfigurationMaps, the outputs from android.scaler.streamConfigurationMapMaximumResolution and android.scaler.streamConfigurationMap must not be mixed in the same CaptureRequest. In other words, these outputs are exclusive to each other. This key does not need to be set for reprocess requests. This key will be be present on devices supporting the CameraMetadata.REQUEST_AVAILABLE_CAPABILITIES_ULTRA_HIGH_RESOLUTION_SENSOR capability. It may also be present on devices which do not support the aforementioned capability. In that case:

Possible values:

Optional - The value for this key may be null on some devices.

SENSOR_RAW_BINNING_FACTOR_USED

Added in API level 31
public static final Key<Boolean> SENSOR_RAW_BINNING_FACTOR_USED

Whether RAW images requested have their bayer pattern as described by android.sensor.info.binningFactor.

This key will only be present in devices advertising the CameraMetadata.REQUEST_AVAILABLE_CAPABILITIES_ULTRA_HIGH_RESOLUTION_SENSOR capability which also advertise REMOSAIC_REPROCESSING capability. On all other devices RAW targets will have a regular bayer pattern.

Optional - The value for this key may be null on some devices.

SENSOR_ROLLING_SHUTTER_SKEW

Added in API level 21
public static final Key<Long> SENSOR_ROLLING_SHUTTER_SKEW

Duration between the start of exposure for the first row of the image sensor, and the start of exposure for one past the last row of the image sensor.

This is the exposure time skew between the first and (last+1) row exposure start times. The first row and the last row are the first and last rows inside of the android.sensor.info.activeArraySize.

For typical camera sensors that use rolling shutters, this is also equivalent to the frame readout time.

If the image sensor is operating in a binned or cropped mode due to the current output target resolutions, it's possible this skew is reported to be larger than the exposure time, for example, since it is based on the full array even if a partial array is read out. Be sure to scale the number to cover the section of the sensor actually being used for the outputs you care about. So if your output covers N rows of the active array of height H, scale this value by N/H to get the total skew for that viewport.

Note: Prior to Android 11, this field was described as measuring duration from first to last row of the image sensor, which is not equal to the frame readout time for a rolling shutter sensor. Implementations generally reported the latter value, so to resolve the inconsistency, the description has been updated to range from (first, last+1) row exposure start, instead.

Units: Nanoseconds

Range of valid values:
>= 0 and < StreamConfigurationMap.getOutputMinFrameDuration(int, Size).

Optional - The value for this key may be null on some devices.

Limited capability - Present on all camera devices that report being at least HARDWARE_LEVEL_LIMITED devices in the android.info.supportedHardwareLevel key

SENSOR_SENSITIVITY

Added in API level 21
public static final Key<Integer> SENSOR_SENSITIVITY

The amount of gain applied to sensor data before processing.

The sensitivity is the standard ISO sensitivity value, as defined in ISO 12232:2006.

The sensitivity must be within android.sensor.info.sensitivityRange, and if if it less than android.sensor.maxAnalogSensitivity, the camera device is guaranteed to use only analog amplification for applying the gain.

If the camera device cannot apply the exact sensitivity requested, it will reduce the gain to the nearest supported value. The final sensitivity used will be available in the output capture result.

This control is only effective if android.control.aeMode or android.control.mode is set to OFF; otherwise the auto-exposure algorithm will override this value.

Note that for devices supporting postRawSensitivityBoost, the total sensitivity applied to the final processed image is the combination of android.sensor.sensitivity and android.control.postRawSensitivityBoost. In case the application uses the sensor sensitivity from last capture result of an auto request for a manual request, in order to achieve the same brightness in the output image, the application should also set postRawSensitivityBoost.

Units: ISO arithmetic units

Range of valid values:
android.sensor.info.sensitivityRange

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

SENSOR_TEST_PATTERN_DATA

Added in API level 21
public static final Key<int[]> SENSOR_TEST_PATTERN_DATA

A pixel [R, G_even, G_odd, B] that supplies the test pattern when android.sensor.testPatternMode is SOLID_COLOR.

Each color channel is treated as an unsigned 32-bit integer. The camera device then uses the most significant X bits that correspond to how many bits are in its Bayer raw sensor output.

For example, a sensor with RAW10 Bayer output would use the 10 most significant bits from each color channel.

Optional - The value for this key may be null on some devices.

SENSOR_TEST_PATTERN_MODE

Added in API level 21
public static final Key<Integer> SENSOR_TEST_PATTERN_MODE

When enabled, the sensor sends a test pattern instead of doing a real exposure from the camera.

When a test pattern is enabled, all manual sensor controls specified by android.sensor.* will be ignored. All other controls should work as normal.

For example, if manual flash is enabled, flash firing should still occur (and that the test pattern remain unmodified, since the flash would not actually affect it).

Defaults to OFF.

Possible values:

Available values for this device:
android.sensor.availableTestPatternModes

Optional - The value for this key may be null on some devices.

SENSOR_TIMESTAMP

Added in API level 21
public static final Key<Long> SENSOR_TIMESTAMP

Time at start of exposure of first row of the image sensor active array, in nanoseconds.

The timestamps are also included in all image buffers produced for the same capture, and will be identical on all the outputs.

When android.sensor.info.timestampSource == UNKNOWN, the timestamps measure time since an unspecified starting point, and are monotonically increasing. They can be compared with the timestamps for other captures from the same camera device, but are not guaranteed to be comparable to any other time source.

When android.sensor.info.timestampSource == REALTIME, the timestamps measure time in the same timebase as SystemClock.elapsedRealtimeNanos(), and they can be compared to other timestamps from other subsystems that are using that base.

For reprocessing, the timestamp will match the start of exposure of the input image, i.e. timestamp in the TotalCaptureResult that was used to create the reprocess capture request.

Units: Nanoseconds

Range of valid values:
> 0

This key is available on all devices.

SHADING_MODE

Added in API level 21
public static final Key<Integer> SHADING_MODE

Quality of lens shading correction applied to the image data.

When set to OFF mode, no lens shading correction will be applied by the camera device, and an identity lens shading map data will be provided if android.statistics.lensShadingMapMode == ON. For example, for lens shading map with size of [ 4, 3 ], the output android.statistics.lensShadingCorrectionMap for this case will be an identity map shown below:

[ 1.0, 1.0, 1.0, 1.0,  1.0, 1.0, 1.0, 1.0,
  1.0, 1.0, 1.0, 1.0,  1.0, 1.0, 1.0, 1.0,
  1.0, 1.0, 1.0, 1.0,  1.0, 1.0, 1.0, 1.0,
  1.0, 1.0, 1.0, 1.0,  1.0, 1.0, 1.0, 1.0,
  1.0, 1.0, 1.0, 1.0,  1.0, 1.0, 1.0, 1.0,
  1.0, 1.0, 1.0, 1.0,  1.0, 1.0, 1.0, 1.0 ]
 

When set to other modes, lens shading correction will be applied by the camera device. Applications can request lens shading map data by setting android.statistics.lensShadingMapMode to ON, and then the camera device will provide lens shading map data in android.statistics.lensShadingCorrectionMap; the returned shading map data will be the one applied by the camera device for this capture request.

The shading map data may depend on the auto-exposure (AE) and AWB statistics, therefore the reliability of the map data may be affected by the AE and AWB algorithms. When AE and AWB are in AUTO modes(android.control.aeMode != OFF and android.control.awbMode != OFF), to get best results, it is recommended that the applications wait for the AE and AWB to be converged before using the returned shading map data.

Possible values:

Available values for this device:
android.shading.availableModes

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

STATISTICS_FACES

Added in API level 21
public static final Key<Face[]> STATISTICS_FACES

List of the faces detected through camera face detection in this capture.

Only available if android.statistics.faceDetectMode != OFF.

This key is available on all devices.

STATISTICS_FACE_DETECT_MODE

Added in API level 21
public static final Key<Integer> STATISTICS_FACE_DETECT_MODE

Operating mode for the face detector unit.

Whether face detection is enabled, and whether it should output just the basic fields or the full set of fields.

Possible values:

Available values for this device:
android.statistics.info.availableFaceDetectModes

This key is available on all devices.

STATISTICS_HOT_PIXEL_MAP

Added in API level 21
public static final Key<Point[]> STATISTICS_HOT_PIXEL_MAP

List of (x, y) coordinates of hot/defective pixels on the sensor.

A coordinate (x, y) must lie between (0, 0), and (width - 1, height - 1) (inclusive), which are the top-left and bottom-right of the pixel array, respectively. The width and height dimensions are given in android.sensor.info.pixelArraySize. This may include hot pixels that lie outside of the active array bounds given by android.sensor.info.activeArraySize.

Range of valid values:

n <= number of pixels on the sensor. The (x, y) coordinates must be bounded by android.sensor.info.pixelArraySize.

Optional - The value for this key may be null on some devices.

STATISTICS_HOT_PIXEL_MAP_MODE

Added in API level 21
public static final Key<Boolean> STATISTICS_HOT_PIXEL_MAP_MODE

Operating mode for hot pixel map generation.

If set to true, a hot pixel map is returned in android.statistics.hotPixelMap. If set to false, no hot pixel map will be returned.

Range of valid values:
android.statistics.info.availableHotPixelMapModes

Optional - The value for this key may be null on some devices.

STATISTICS_LENS_INTRINSICS_SAMPLES

Added in API level 35
public static final Key<LensIntrinsicsSample[]> STATISTICS_LENS_INTRINSICS_SAMPLES

An array of intra-frame lens intrinsic samples.

Contains an array of intra-frame android.lens.intrinsicCalibration updates. This must not be confused or compared to android.statistics.oisSamples. Although OIS could be the main driver, all relevant factors such as focus distance and optical zoom must also be included. Do note that OIS samples must not be applied on top of the lens intrinsic samples. Support for this capture result can be queried via CameraCharacteristics.getAvailableCaptureResultKeys(). If available, clients can expect multiple samples per capture result. The specific amount will depend on current frame duration and sampling rate. Generally a sampling rate greater than or equal to 200Hz is considered sufficient for high quality results.

Optional - The value for this key may be null on some devices.

STATISTICS_LENS_SHADING_CORRECTION_MAP

Added in API level 21
public static final Key<LensShadingMap> STATISTICS_LENS_SHADING_CORRECTION_MAP

The shading map is a low-resolution floating-point map that lists the coefficients used to correct for vignetting, for each Bayer color channel.

The map provided here is the same map that is used by the camera device to correct both color shading and vignetting for output non-RAW images.

When there is no lens shading correction applied to RAW output images (android.sensor.info.lensShadingApplied == false), this map is the complete lens shading correction map; when there is some lens shading correction applied to the RAW output image (android.sensor.info.lensShadingApplied== true), this map reports the remaining lens shading correction map that needs to be applied to get shading corrected images that match the camera device's output for non-RAW formats.

Therefore, whatever the value of lensShadingApplied is, the lens shading map should always be applied to RAW images if the goal is to match the shading appearance of processed (non-RAW) images.

For a complete shading correction map, the least shaded section of the image will have a gain factor of 1; all other sections will have gains above 1.

When android.colorCorrection.mode = TRANSFORM_MATRIX, the map will take into account the colorCorrection settings.

The shading map is for the entire active pixel array, and is not affected by the crop region specified in the request. Each shading map entry is the value of the shading compensation map over a specific pixel on the sensor. Specifically, with a (N x M) resolution shading map, and an active pixel array size (W x H), shading map entry (x,y) \u03f5 (0 ... N-1, 0 ... M-1) is the value of the shading map at pixel ( ((W-1)/(N-1)) * x, ((H-1)/(M-1)) * y) for the four color channels. The map is assumed to be bilinearly interpolated between the sample points.

The channel order is [R, Geven, Godd, B], where Geven is the green channel for the even rows of a Bayer pattern, and Godd is the odd rows. The shading map is stored in a fully interleaved format.

The shading map will generally have on the order of 30-40 rows and columns, and will be smaller than 64x64.

As an example, given a very small map defined as:

width,height = [ 4, 3 ]
 values =
 [ 1.3, 1.2, 1.15, 1.2,  1.2, 1.2, 1.15, 1.2,
     1.1, 1.2, 1.2, 1.2,  1.3, 1.2, 1.3, 1.3,
   1.2, 1.2, 1.25, 1.1,  1.1, 1.1, 1.1, 1.0,
     1.0, 1.0, 1.0, 1.0,  1.2, 1.3, 1.25, 1.2,
   1.3, 1.2, 1.2, 1.3,   1.2, 1.15, 1.1, 1.2,
     1.2, 1.1, 1.0, 1.2,  1.3, 1.15, 1.2, 1.3 ]
 

The low-resolution scaling map images for each channel are (displayed using nearest-neighbor interpolation):

Red lens shading map Green (even rows) lens shading map Green (odd rows) lens shading map Blue lens shading map

As a visualization only, inverting the full-color map to recover an image of a gray wall (using bicubic interpolation for visual quality) as captured by the sensor gives:

Image of a uniform white wall (inverse shading map)

For a MONOCHROME camera, all of the 2x2 channels must have the same values. An example shading map for such a camera is defined as:

android.lens.info.shadingMapSize = [ 4, 3 ]
 android.statistics.lensShadingMap =
 [ 1.3, 1.3, 1.3, 1.3,  1.2, 1.2, 1.2, 1.2,
     1.1, 1.1, 1.1, 1.1,  1.3, 1.3, 1.3, 1.3,
   1.2, 1.2, 1.2, 1.2,  1.1, 1.1, 1.1, 1.1,
     1.0, 1.0, 1.0, 1.0,  1.2, 1.2, 1.2, 1.2,
   1.3, 1.3, 1.3, 1.3,   1.2, 1.2, 1.2, 1.2,
     1.2, 1.2, 1.2, 1.2,  1.3, 1.3, 1.3, 1.3 ]
 

Range of valid values:
Each gain factor is >= 1

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

STATISTICS_LENS_SHADING_MAP_MODE

Added in API level 21
public static final Key<Integer> STATISTICS_LENS_SHADING_MAP_MODE

Whether the camera device will output the lens shading map in output result metadata.

When set to ON, android.statistics.lensShadingMap will be provided in the output result metadata.

ON is always supported on devices with the RAW capability.

Possible values:

Available values for this device:
android.statistics.info.availableLensShadingMapModes

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

STATISTICS_OIS_DATA_MODE

Added in API level 28
public static final Key<Integer> STATISTICS_OIS_DATA_MODE

A control for selecting whether optical stabilization (OIS) position information is included in output result metadata.

Since optical image stabilization generally involves motion much faster than the duration of individual image exposure, multiple OIS samples can be included for a single capture result. For example, if the OIS reporting operates at 200 Hz, a typical camera operating at 30fps may have 6-7 OIS samples per capture result. This information can be combined with the rolling shutter skew to account for lens motion during image exposure in post-processing algorithms.

Possible values:

Available values for this device:
android.statistics.info.availableOisDataModes

Optional - The value for this key may be null on some devices.

STATISTICS_OIS_SAMPLES

Added in API level 28
public static final Key<OisSample[]> STATISTICS_OIS_SAMPLES

An array of optical stabilization (OIS) position samples.

Each OIS sample contains the timestamp and the amount of shifts in x and y direction, in pixels, of the OIS sample.

A positive value for a shift in x direction is a shift from left to right in the pre-correction active array coordinate system. For example, if the optical center is (1000, 500) in pre-correction active array coordinates, a shift of (3, 0) puts the new optical center at (1003, 500).

A positive value for a shift in y direction is a shift from top to bottom in pre-correction active array coordinate system. For example, if the optical center is (1000, 500) in active array coordinates, a shift of (0, 5) puts the new optical center at (1000, 505).

The OIS samples are not affected by whether lens distortion correction is enabled (on supporting devices). They are always reported in pre-correction active array coordinates, since the scaling of OIS shifts would depend on the specific spot on the sensor the shift is needed.

Optional - The value for this key may be null on some devices.

STATISTICS_SCENE_FLICKER

Added in API level 21
public static final Key<Integer> STATISTICS_SCENE_FLICKER

The camera device estimated scene illumination lighting frequency.

Many light sources, such as most fluorescent lights, flicker at a rate that depends on the local utility power standards. This flicker must be accounted for by auto-exposure routines to avoid artifacts in captured images. The camera device uses this entry to tell the application what the scene illuminant frequency is.

When manual exposure control is enabled (android.control.aeMode == OFF or android.control.mode == OFF), the android.control.aeAntibandingMode doesn't perform antibanding, and the application can ensure it selects exposure times that do not cause banding issues by looking into this metadata field. See android.control.aeAntibandingMode for more details.

Reports NONE if there doesn't appear to be flickering illumination.

Possible values:

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

TONEMAP_CURVE

Added in API level 21
public static final Key<TonemapCurve> TONEMAP_CURVE

Tonemapping / contrast / gamma curve to use when android.tonemap.mode is CONTRAST_CURVE.

The tonemapCurve consist of three curves for each of red, green, and blue channels respectively. The following example uses the red channel as an example. The same logic applies to green and blue channel. Each channel's curve is defined by an array of control points:

curveRed =
   [ P0(in, out), P1(in, out), P2(in, out), P3(in, out), ..., PN(in, out) ]
 2 <= N <= android.tonemap.maxCurvePoints

These are sorted in order of increasing Pin; it is always guaranteed that input values 0.0 and 1.0 are included in the list to define a complete mapping. For input values between control points, the camera device must linearly interpolate between the control points.

Each curve can have an independent number of points, and the number of points can be less than max (that is, the request doesn't have to always provide a curve with number of points equivalent to android.tonemap.maxCurvePoints).

For devices with MONOCHROME capability, all three channels must have the same set of control points.

A few examples, and their corresponding graphical mappings; these only specify the red channel and the precision is limited to 4 digits, for conciseness.

Linear mapping:

curveRed = [ (0, 0), (1.0, 1.0) ]
 

Linear mapping curve

Invert mapping:

curveRed = [ (0, 1.0), (1.0, 0) ]
 

Inverting mapping curve

Gamma 1/2.2 mapping, with 16 control points:

curveRed = [
   (0.0000, 0.0000), (0.0667, 0.2920), (0.1333, 0.4002), (0.2000, 0.4812),
   (0.2667, 0.5484), (0.3333, 0.6069), (0.4000, 0.6594), (0.4667, 0.7072),
   (0.5333, 0.7515), (0.6000, 0.7928), (0.6667, 0.8317), (0.7333, 0.8685),
   (0.8000, 0.9035), (0.8667, 0.9370), (0.9333, 0.9691), (1.0000, 1.0000) ]
 

Gamma = 1/2.2 tonemapping curve

Standard sRGB gamma mapping, per IEC 61966-2-1:1999, with 16 control points:

curveRed = [
   (0.0000, 0.0000), (0.0667, 0.2864), (0.1333, 0.4007), (0.2000, 0.4845),
   (0.2667, 0.5532), (0.3333, 0.6125), (0.4000, 0.6652), (0.4667, 0.7130),
   (0.5333, 0.7569), (0.6000, 0.7977), (0.6667, 0.8360), (0.7333, 0.8721),
   (0.8000, 0.9063), (0.8667, 0.9389), (0.9333, 0.9701), (1.0000, 1.0000) ]
 

sRGB tonemapping curve

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

TONEMAP_GAMMA

Added in API level 23
public static final Key<Float> TONEMAP_GAMMA

Tonemapping curve to use when android.tonemap.mode is GAMMA_VALUE

The tonemap curve will be defined the following formula:

  • OUT = pow(IN, 1.0 / gamma)

where IN and OUT is the input pixel value scaled to range [0.0, 1.0], pow is the power function and gamma is the gamma value specified by this key.

The same curve will be applied to all color channels. The camera device may clip the input gamma value to its supported range. The actual applied value will be returned in capture result.

The valid range of gamma value varies on different devices, but values within [1.0, 5.0] are guaranteed not to be clipped.

Optional - The value for this key may be null on some devices.

TONEMAP_MODE

Added in API level 21
public static final Key<Integer> TONEMAP_MODE

High-level global contrast/gamma/tonemapping control.

When switching to an application-defined contrast curve by setting android.tonemap.mode to CONTRAST_CURVE, the curve is defined per-channel with a set of (in, out) points that specify the mapping from input high-bit-depth pixel value to the output low-bit-depth value. Since the actual pixel ranges of both input and output may change depending on the camera pipeline, the values are specified by normalized floating-point numbers.

More-complex color mapping operations such as 3D color look-up tables, selective chroma enhancement, or other non-linear color transforms will be disabled when android.tonemap.mode is CONTRAST_CURVE.

When using either FAST or HIGH_QUALITY, the camera device will emit its own tonemap curve in android.tonemap.curve. These values are always available, and as close as possible to the actually used nonlinear/nonglobal transforms.

If a request is sent with CONTRAST_CURVE with the camera device's provided curve in FAST or HIGH_QUALITY, the image's tonemap will be roughly the same.

Possible values:

Available values for this device:
android.tonemap.availableToneMapModes

Optional - The value for this key may be null on some devices.

Full capability - Present on all camera devices that report being HARDWARE_LEVEL_FULL devices in the android.info.supportedHardwareLevel key

TONEMAP_PRESET_CURVE

Added in API level 23
public static final Key<Integer> TONEMAP_PRESET_CURVE

Tonemapping curve to use when android.tonemap.mode is PRESET_CURVE

The tonemap curve will be defined by specified standard.

sRGB (approximated by 16 control points):

sRGB tonemapping curve

Rec. 709 (approximated by 16 control points):

Rec. 709 tonemapping curve

Note that above figures show a 16 control points approximation of preset curves. Camera devices may apply a different approximation to the curve.

Possible values:

Optional - The value for this key may be null on some devices.

Public methods

get

Added in API level 21
public T get (Key<T> key)

Get a capture result field value.

The field definitions can be found in CaptureResult.

Querying the value for the same key more than once will return a value which is equal to the previous queried value.

Parameters
key Key: The result field to read.

Returns
T The value of that key, or null if the field is not set.

Throws
IllegalArgumentException if the key was not valid

getCameraId

Added in API level 31
public String getCameraId ()

Get the camera ID of the camera that produced this capture result. For a logical multi-camera, the ID may be the logical or the physical camera ID, depending on whether the capture result was obtained from TotalCaptureResult#getPhysicalCameraResults or not.

Returns
String The camera ID for the camera that produced this capture result. This value cannot be null.

getFrameNumber

Added in API level 21
public long getFrameNumber ()

Get the frame number associated with this result.

Whenever a request has been processed, regardless of failure or success, it gets a unique frame number assigned to its future result/failure.

For the same type of request (capturing from the camera device or reprocessing), this value monotonically increments, starting with 0, for every new result or failure and the scope is the lifetime of the CameraDevice. Between different types of requests, the frame number may not monotonically increment. For example, the frame number of a newer reprocess result may be smaller than the frame number of an older result of capturing new images from the camera device, but the frame number of a newer reprocess result will never be smaller than the frame number of an older reprocess result.

Returns
long The frame number

getKeys

Added in API level 21
public List<Key<?>> getKeys ()

Returns a list of the keys contained in this map.

The list returned is not modifiable, so any attempts to modify it will throw a UnsupportedOperationException.

All values retrieved by a key from this list with #get are guaranteed to be non-null. Each key is only listed once in the list. The order of the keys is undefined.

Returns
List<Key<?>> This value cannot be null.

getRequest

Added in API level 21
public CaptureRequest getRequest ()

Get the request associated with this result.

Whenever a request has been fully or partially captured, with CameraCaptureSession.CaptureCallback#onCaptureCompleted or CameraCaptureSession.CaptureCallback#onCaptureProgressed, the result's getRequest() will return that request.

For example,

cameraDevice.capture(someRequest, new CaptureCallback() {
     @Override
     void onCaptureCompleted(CaptureRequest myRequest, CaptureResult myResult) {
         assert(myResult.getRequest.equals(myRequest) == true);
     }
 }, null);
 

Returns
CaptureRequest The request associated with this result. Never null.

getSequenceId

Added in API level 21
public int getSequenceId ()

The sequence ID for this failure that was returned by the CameraCaptureSession#capture family of functions.

The sequence ID is a unique monotonically increasing value starting from 0, incremented every time a new group of requests is submitted to the CameraDevice.

Returns
int int The ID for the sequence of requests that this capture result is a part of