Android APIs
public class

CaptureResult

extends CameraMetadata<Key<?>>
java.lang.Object
   ↳ android.hardware.camera2.CameraMetadata<android.hardware.camera2.CaptureResult.Key<?>>
     ↳ android.hardware.camera2.CaptureResult
Known Direct Subclasses

Class Overview

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 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 get(CaptureResult.Key)
[Expand]
Inherited Constants
From class android.hardware.camera2.CameraMetadata
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_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<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<Integer> CONTROL_MODE

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

public static final Key<Integer> CONTROL_SCENE_MODE

Control for which scene mode is currently active.

public static final Key<Integer> CONTROL_VIDEO_STABILIZATION_MODE

Whether video stabilization is active.

public static final Key<Integer> EDGE_MODE

Operation mode for edge enhancement.

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> 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_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

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

public static final Key<Integer> LENS_STATE

Current lens status.

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<Long> SENSOR_EXPOSURE_TIME

Duration each pixel is exposed to light.

public static final Key<Long> SENSOR_FRAME_DURATION

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

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<Long> SENSOR_ROLLING_SHUTTER_SKEW

Duration between the start of first row exposure and the start of last row exposure.

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<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_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.
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 capture(CaptureRequest, CameraCaptureSession.CaptureCallback, Handler) family of functions.
[Expand]
Inherited Methods
From class android.hardware.camera2.CameraMetadata
From class java.lang.Object

Fields

public static final Key<Boolean> BLACK_LEVEL_LOCK

Added in API level 21

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 - This value 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

public static final Key<Integer> COLOR_CORRECTION_ABERRATION_MODE

Added in API level 21

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.

public static final Key<RggbChannelVector> COLOR_CORRECTION_GAINS

Added in API level 21

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 - This value 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

public static final Key<Integer> COLOR_CORRECTION_MODE

Added in API level 21

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 FULL 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 - This value 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

public static final Key<ColorSpaceTransform> COLOR_CORRECTION_TRANSFORM

Added in API level 21

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 - This value 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

public static final Key<Integer> CONTROL_AE_ANTIBANDING_MODE

Added in API level 21

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.

public static final Key<Integer> CONTROL_AE_EXPOSURE_COMPENSATION

Added in API level 21

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.

public static final Key<Boolean> CONTROL_AE_LOCK

Added in API level 21

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.

public static final Key<Integer> CONTROL_AE_MODE

Added in API level 21

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.

public static final Key<Integer> CONTROL_AE_PRECAPTURE_TRIGGER

Added in API level 21

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 - This value 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

public static final Key<MeteringRectangle[]> CONTROL_AE_REGIONS

Added in API level 21

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.

The coordinate system is based on the active pixel array, 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.

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.

Units: Pixel coordinates within android.sensor.info.activeArraySize

Range of valid values:
Coordinates must be between [(0,0), (width, height)) of android.sensor.info.activeArraySize

Optional - This value may be null on some devices.

public static final Key<Integer> CONTROL_AE_STATE

Added in API level 21

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

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 sequenceis 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 - This value 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

public static final Key<Range<Integer>> CONTROL_AE_TARGET_FPS_RANGE

Added in API level 21

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.

Units: Frames per second (FPS)

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

This key is available on all devices.

public static final Key<Integer> CONTROL_AF_MODE

Added in API level 21

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.

public static final Key<MeteringRectangle[]> CONTROL_AF_REGIONS

Added in API level 21

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.

The coordinate system is based on the active pixel array, 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.

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.

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.

Units: Pixel coordinates within android.sensor.info.activeArraySize

Range of valid values:
Coordinates must be between [(0,0), (width, height)) of android.sensor.info.activeArraySize

Optional - This value may be null on some devices.

public static final Key<Integer> CONTROL_AF_STATE

Added in API level 21

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.

public static final Key<Integer> CONTROL_AF_TRIGGER

Added in API level 21

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.

public static final Key<Boolean> CONTROL_AWB_LOCK

Added in API level 21

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.

See Also

public static final Key<Integer> CONTROL_AWB_MODE

Added in API level 21

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 ON 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 recommened 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.

public static final Key<MeteringRectangle[]> CONTROL_AWB_REGIONS

Added in API level 21

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.

The coordinate system is based on the active pixel array, 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.

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.

Units: Pixel coordinates within android.sensor.info.activeArraySize

Range of valid values:
Coordinates must be between [(0,0), (width, height)) of android.sensor.info.activeArraySize

Optional - This value may be null on some devices.

public static final Key<Integer> CONTROL_AWB_STATE

Added in API level 21

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 - This value 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

public static final Key<Integer> CONTROL_CAPTURE_INTENT

Added in API level 21

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.

ZERO_SHUTTER_LAG will be supported if android.request.availableCapabilities contains PRIVATE_REPROCESSING or YUV_REPROCESSING. MANUAL will be supported if android.request.availableCapabilities contains MANUAL_SENSOR. Other intent values are always supported.

Possible values:

This key is available on all devices.

public static final Key<Integer> CONTROL_EFFECT_MODE

Added in API level 21

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.

public static final Key<Integer> CONTROL_MODE

Added in API level 21

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, the individual controls in android.control.* are mostly disabled, and the camera device implements one of the scene mode settings (such as ACTION, SUNSET, or PARTY) 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.

public static final Key<Integer> CONTROL_SCENE_MODE

Added in API level 21

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.

public static final Key<Integer> CONTROL_VIDEO_STABILIZATION_MODE

Added in API level 21

Whether video stabilization is active.

Video stabilization automatically translates and scales 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.

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.

Possible values:

This key is available on all devices.

public static final Key<Integer> EDGE_MODE

Added in API level 21

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 - This value 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

public static final Key<Integer> FLASH_MODE

Added in API level 21

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.

public static final Key<Integer> FLASH_STATE

Added in API level 21

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 - This value 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

public static final Key<Integer> HOT_PIXEL_MODE

Added in API level 21

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 - This value may be null on some devices.

public static final Key<Location> JPEG_GPS_LOCATION

Added in API level 21

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 key is available on all devices.

public static final Key<Integer> JPEG_ORIENTATION

Added in API level 21

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.

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, 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;
 }
 

Units: Degrees in multiples of 90

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

This key is available on all devices.

public static final Key<Byte> JPEG_QUALITY

Added in API level 21

Compression quality of the final JPEG image.

85-95 is typical usage range.

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

This key is available on all devices.

public static final Key<Byte> JPEG_THUMBNAIL_QUALITY

Added in API level 21

Compression quality of JPEG thumbnail.

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

This key is available on all devices.

public static final Key<Size> JPEG_THUMBNAIL_SIZE

Added in API level 21

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 thumnail 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.

Range of valid values:
android.jpeg.availableThumbnailSizes

This key is available on all devices.

public static final Key<Float> LENS_APERTURE

Added in API level 21

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 - This value 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

public static final Key<Float> LENS_FILTER_DENSITY

Added in API level 21

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 - This value 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

public static final Key<Float> LENS_FOCAL_LENGTH

Added in API level 21

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 will not be supported on most devices.

Units: Millimeters

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

This key is available on all devices.

public static final Key<Float> LENS_FOCUS_DISTANCE

Added in API level 21

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 - This value 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

public static final Key<Pair<FloatFloat>> LENS_FOCUS_RANGE

Added in API level 21

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 - This value 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

public static final Key<float[]> LENS_INTRINSIC_CALIBRATION

Added in API level 23

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, respective) to calculate the complete transform from world coordinates to pixel coordinates:

P = [ K 0   * [ R t
      0 1 ]     0 1 ]
 

and 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.radialDistortion 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.

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

Optional - This value may be null on some devices.

public static final Key<Integer> LENS_OPTICAL_STABILIZATION_MODE

Added in API level 21

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.

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 - This value 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

public static final Key<float[]> LENS_POSE_ROTATION

Added in API level 23

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

The four coefficients that describe the quarternion 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 quarternion 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 quarternion, 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.

Units: Quarternion coefficients

Optional - This value may be null on some devices.

public static final Key<float[]> LENS_POSE_TRANSLATION

Added in API level 23

Position of the camera optical center.

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

To transform a pixel coordinates between two cameras facing the same direction, first the source camera android.lens.radialDistortion 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.

Units: Meters

Optional - This value may be null on some devices.

public static final Key<float[]> LENS_RADIAL_DISTORTION

Added in API level 23

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 - This value may be null on some devices.

public static final Key<Integer> LENS_STATE

Added in API level 21

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 changable 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 - This value 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

public static final Key<Integer> NOISE_REDUCTION_MODE

Added in API level 21

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 - This value 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

public static final Key<Float> REPROCESS_EFFECTIVE_EXPOSURE_FACTOR

Added in API level 23

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 - This value 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

public static final Key<Byte> REQUEST_PIPELINE_DEPTH

Added in API level 21

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.

public static final Key<Rect> SCALER_CROP_REGION

Added in API level 21

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

This control can be used to implement digital zoom.

The crop region coordinate system is based off android.sensor.info.activeArraySize, with (0, 0) being the top-left corner of the sensor 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 applied after the RAW to other color space (e.g. YUV) conversion. Since raw streams (e.g. RAW16) don't have the conversion stage, they are not croppable. The crop region will be ignored by raw streams.

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.

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.

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.

Units: Pixel coordinates relative to android.sensor.info.activeArraySize

This key is available on all devices.

public static final Key<Long> SENSOR_EXPOSURE_TIME

Added in API level 21

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 - This value 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

public static final Key<Long> SENSOR_FRAME_DURATION

Added in API level 21

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

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 the android.scaler.streamConfigurationMap field using 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 android.scaler.streamConfigurationMap using 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 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 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.

Units: Nanoseconds

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

Optional - This value 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

public static final Key<Float> SENSOR_GREEN_SPLIT

Added in API level 21

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 usuable image (>20% divergence).

Range of valid values:

>= 0

Optional - This value may be null on some devices.

public static final Key<Rational[]> SENSOR_NEUTRAL_COLOR_POINT

Added in API level 21

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.

Optional - This value may be null on some devices.

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

Added in API level 21

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 coeffiecients for that channel.

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

Optional - This value may be null on some devices.

public static final Key<Long> SENSOR_ROLLING_SHUTTER_SKEW

Added in API level 21

Duration between the start of first row exposure and the start of last row exposure.

This is the exposure time skew between the first and last 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.

Units: Nanoseconds

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

Optional - This value 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

public static final Key<Integer> SENSOR_SENSITIVITY

Added in API level 21

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.

Units: ISO arithmetic units

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

Optional - This value 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

public static final Key<int[]> SENSOR_TEST_PATTERN_DATA

Added in API level 21

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 - This value may be null on some devices.

public static final Key<Integer> SENSOR_TEST_PATTERN_MODE

Added in API level 21

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 - This value may be null on some devices.

public static final Key<Long> SENSOR_TIMESTAMP

Added in API level 21

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 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. the 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.

public static final Key<Integer> SHADING_MODE

Added in API level 21

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 - This value 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

public static final Key<Face[]> STATISTICS_FACES

Added in API level 21

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.

public static final Key<Integer> STATISTICS_FACE_DETECT_MODE

Added in API level 21

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.

public static final Key<Point[]> STATISTICS_HOT_PIXEL_MAP

Added in API level 21

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 - This value may be null on some devices.

public static final Key<Boolean> STATISTICS_HOT_PIXEL_MAP_MODE

Added in API level 21

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 - This value may be null on some devices.

public static final Key<LensShadingMap> STATISTICS_LENS_SHADING_CORRECTION_MAP

Added in API level 21

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 least shaded section of the image should have a gain factor of 1; all other sections should have gains above 1.

When android.colorCorrection.mode = TRANSFORM_MATRIX, the map must 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) ϵ (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 should have on the order of 30-40 rows and columns, and must 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)

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

Optional - This value 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

public static final Key<Integer> STATISTICS_LENS_SHADING_MAP_MODE

Added in API level 21

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 - This value 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

public static final Key<Integer> STATISTICS_SCENE_FLICKER

Added in API level 21

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 - This value 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

public static final Key<TonemapCurve> TONEMAP_CURVE

Added in API level 21

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).

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 - This value 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

public static final Key<Float> TONEMAP_GAMMA

Added in API level 23

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 - This value may be null on some devices.

See Also

public static final Key<Integer> TONEMAP_MODE

Added in API level 21

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 - This value 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

public static final Key<Integer> TONEMAP_PRESET_CURVE

Added in API level 23

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 - This value may be null on some devices.

Public Methods

public T get (Key<T> key)

Added in API level 21

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 The result field to read.
Returns
  • The value of that key, or null if the field is not set.
Throws
IllegalArgumentException if the key was not valid

public long getFrameNumber ()

Added in API level 21

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
  • The frame number

public List<Key<?>> getKeys ()

Added in API level 21

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 of the keys contained in this map.

public CaptureRequest getRequest ()

Added in API level 21

Get the request associated with this result.

Whenever a request has been fully or partially captured, with onCaptureCompleted(CameraCaptureSession, CaptureRequest, TotalCaptureResult) or onCaptureProgressed(CameraCaptureSession, CaptureRequest, CaptureResult), 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
  • The request associated with this result. Never null.

public int getSequenceId ()

Added in API level 21

The sequence ID for this failure that was returned by the capture(CaptureRequest, CameraCaptureSession.CaptureCallback, Handler) 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 The ID for the sequence of requests that this capture result is a part of