CaptureResult

BLACK_LEVEL_LOCK

public static final Key<Boolean> BLACK_LEVEL_LOCK

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

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

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

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

COLOR_CORRECTION_ABERRATION_MODE

public static final Key<Integer> COLOR_CORRECTION_ABERRATION_MODE

Mode of operation for the chromatic aberration correction algorithm.

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

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

LEGACY devices will always be in FAST mode.

Possible values:

• OFF
• FAST
• HIGH_QUALITY

Available values for this device:
android.colorCorrection.availableAberrationModes

This key is available on all devices.

COLOR_CORRECTION_GAINS

public static final Key<RggbChannelVector> COLOR_CORRECTION_GAINS

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

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

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

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

Units: Unitless gain factors

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

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

COLOR_CORRECTION_MODE

public static final Key<Integer> COLOR_CORRECTION_MODE

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

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

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

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

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

The expected processing pipeline is as follows:

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:

• TRANSFORM_MATRIX
• FAST
• HIGH_QUALITY

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

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

COLOR_CORRECTION_TRANSFORM

public static final Key<ColorSpaceTransform> COLOR_CORRECTION_TRANSFORM

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

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

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

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

Units: Unitless scale factors

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

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

CONTROL_AE_ANTIBANDING_MODE

public static final Key<Integer> CONTROL_AE_ANTIBANDING_MODE

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

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

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

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

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

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

Possible values:

• OFF
• 50HZ
• 60HZ
• AUTO

Available values for this device:

android.control.aeAvailableAntibandingModes

This key is available on all devices.

CONTROL_AE_EXPOSURE_COMPENSATION

public static final Key<Integer> CONTROL_AE_EXPOSURE_COMPENSATION

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

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

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

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

Units: Compensation steps

Range of valid values:
android.control.aeCompensationRange

This key is available on all devices.

CONTROL_AE_LOCK

public static final Key<Boolean> CONTROL_AE_LOCK

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

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

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

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

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

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

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

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

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

This key is available on all devices.

CONTROL_AE_MODE

public static final Key<Integer> CONTROL_AE_MODE

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

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

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

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

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

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

Possible values:

• OFF
• ON
• ON_AUTO_FLASH
• ON_ALWAYS_FLASH
• ON_AUTO_FLASH_REDEYE
• ON_EXTERNAL_FLASH

Available values for this device:
android.control.aeAvailableModes

This key is available on all devices.

CONTROL_AE_PRECAPTURE_TRIGGER

public static final Key<Integer> CONTROL_AE_PRECAPTURE_TRIGGER

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

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

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

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

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

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

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

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

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

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

Possible values:

• IDLE
• START
• CANCEL

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

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

CONTROL_AE_REGIONS

public static final Key<MeteringRectangle[]> CONTROL_AE_REGIONS

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

CONTROL_AE_STATE

public static final Key<Integer> CONTROL_AE_STATE

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

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

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

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

Below are state transition tables for different AE modes.

When android.control.aeMode is AE_MODE_ON*:

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

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

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

Possible values:

• INACTIVE
• SEARCHING
• CONVERGED
• LOCKED
• FLASH_REQUIRED
• PRECAPTURE

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

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

CONTROL_AE_TARGET_FPS_RANGE

public static final Key<Range<Integer>> CONTROL_AE_TARGET_FPS_RANGE

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

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

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

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

Units: Frames per second (FPS)

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

This key is available on all devices.

CONTROL_AF_MODE

public static final Key<Integer> CONTROL_AF_MODE

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

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

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

Possible values:

• OFF
• AUTO
• MACRO
• CONTINUOUS_VIDEO
• CONTINUOUS_PICTURE
• EDOF

Available values for this device:
android.control.afAvailableModes

This key is available on all devices.

CONTROL_AF_REGIONS

public static final Key<MeteringRectangle[]> CONTROL_AF_REGIONS

List of metering areas to use for auto-focus.

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

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

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

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

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

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

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

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

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

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

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

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

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

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

CONTROL_AF_SCENE_CHANGE

public static final Key<Integer> CONTROL_AF_SCENE_CHANGE

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

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

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

Possible values:

• NOT_DETECTED
• DETECTED

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

CONTROL_AF_STATE

public static final Key<Integer> CONTROL_AF_STATE

Current state of auto-focus (AF) algorithm.

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

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

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

Below are state transition tables for different AF modes.

When android.control.afMode is AF_MODE_OFF or AF_MODE_EDOF:

When android.control.afMode is AF_MODE_AUTO or AF_MODE_MACRO:

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:

When android.control.afMode is AF_MODE_CONTINUOUS_VIDEO:

When android.control.afMode is AF_MODE_CONTINUOUS_PICTURE:

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:

Possible values:

• INACTIVE
• PASSIVE_SCAN
• PASSIVE_FOCUSED
• ACTIVE_SCAN
• FOCUSED_LOCKED
• NOT_FOCUSED_LOCKED
• PASSIVE_UNFOCUSED

This key is available on all devices.

CONTROL_AF_TRIGGER

public static final Key<Integer> CONTROL_AF_TRIGGER

Whether the camera device will trigger autofocus for this request.

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

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

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

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

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

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

Possible values:

• IDLE
• START
• CANCEL

This key is available on all devices.

CONTROL_AUTOFRAMING

public static final Key<Integer> CONTROL_AUTOFRAMING

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

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

Possible values:

• OFF
• ON

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

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

CONTROL_AUTOFRAMING_STATE

public static final Key<Integer> CONTROL_AUTOFRAMING_STATE

Current state of auto-framing.

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

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

Possible values:

• INACTIVE
• FRAMING
• CONVERGED

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

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

CONTROL_AWB_LOCK

public static final Key<Boolean> CONTROL_AWB_LOCK

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

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

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

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

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

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

This key is available on all devices.

CONTROL_AWB_MODE

public static final Key<Integer> CONTROL_AWB_MODE

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

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

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

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

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

Possible values:

• OFF
• AUTO
• INCANDESCENT
• FLUORESCENT
• WARM_FLUORESCENT
• DAYLIGHT
• CLOUDY_DAYLIGHT
• TWILIGHT
• SHADE

Available values for this device:
android.control.awbAvailableModes

This key is available on all devices.

CONTROL_AWB_REGIONS

public static final Key<MeteringRectangle[]> CONTROL_AWB_REGIONS

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

CONTROL_AWB_STATE

public static final Key<Integer> CONTROL_AWB_STATE

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

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

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

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

Below are state transition tables for different AWB modes.

When android.control.awbMode != AWB_MODE_AUTO:

When android.control.awbMode is AWB_MODE_AUTO:

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:

Possible values:

• INACTIVE
• SEARCHING
• CONVERGED
• LOCKED

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

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

CONTROL_CAPTURE_INTENT

public static final Key<Integer> CONTROL_CAPTURE_INTENT

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

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

All intents are supported by all devices, except that:

• 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.
• MOTION_TRACKING will be supported if android.request.availableCapabilities contains MOTION_TRACKING.

Possible values:

• CUSTOM
• PREVIEW
• STILL_CAPTURE
• VIDEO_RECORD
• VIDEO_SNAPSHOT
• ZERO_SHUTTER_LAG
• MANUAL
• MOTION_TRACKING

This key is available on all devices.

CONTROL_EFFECT_MODE

public static final Key<Integer> CONTROL_EFFECT_MODE

A special color effect to apply.

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

Possible values:

• OFF
• MONO
• NEGATIVE
• SOLARIZE
• SEPIA
• POSTERIZE
• WHITEBOARD
• BLACKBOARD
• AQUA

Available values for this device:
android.control.availableEffects

This key is available on all devices.

CONTROL_ENABLE_ZSL

public static final Key<Boolean> CONTROL_ENABLE_ZSL

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

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

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

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

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

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

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

For application-operated ZSL, use CAMERA3_TEMPLATE_ZERO_SHUTTER_LAG template.

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

CONTROL_EXTENDED_SCENE_MODE

public static final Key<Integer> CONTROL_EXTENDED_SCENE_MODE

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

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

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

When set to BOKEH_STILL_CAPTURE mode with PREVIEW capture intent,

• If the camera device has BURST_CAPTURE capability, the frame rate requirement of BURST_CAPTURE must still be met.
• All streams not larger than the maximum streaming dimension for BOKEH_STILL_CAPTURE mode (queried via CameraCharacteristics.CONTROL_AVAILABLE_EXTENDED_SCENE_MODE_CAPABILITIES) will have preview bokeh effect applied.

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

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

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

Possible values:

• DISABLED
• BOKEH_STILL_CAPTURE
• BOKEH_CONTINUOUS

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

CONTROL_LOW_LIGHT_BOOST_STATE

public static final Key<Integer> CONTROL_LOW_LIGHT_BOOST_STATE

Current state of the low light boost AE mode.

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

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

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

The default value will always be 'INACTIVE'.

Possible values:

• INACTIVE
• ACTIVE

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

CONTROL_MODE

public static final Key<Integer> CONTROL_MODE

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

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

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

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

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

Possible values:

• OFF
• AUTO
• USE_SCENE_MODE
• OFF_KEEP_STATE
• USE_EXTENDED_SCENE_MODE

Available values for this device:
android.control.availableModes

This key is available on all devices.

CONTROL_POST_RAW_SENSITIVITY_BOOST

public static final Key<Integer> CONTROL_POST_RAW_SENSITIVITY_BOOST

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

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

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

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

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

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

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

Range of valid values:
android.control.postRawSensitivityBoostRange

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

CONTROL_SCENE_MODE

public static final Key<Integer> CONTROL_SCENE_MODE

Control for which scene mode is currently active.

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

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

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

Possible values:

• DISABLED
• FACE_PRIORITY
• ACTION
• PORTRAIT
• LANDSCAPE
• NIGHT
• NIGHT_PORTRAIT
• THEATRE
• BEACH
• SNOW
• SUNSET
• STEADYPHOTO
• FIREWORKS
• SPORTS
• PARTY
• CANDLELIGHT
• BARCODE
• HIGH_SPEED_VIDEO
• HDR

Available values for this device:
android.control.availableSceneModes

This key is available on all devices.

CONTROL_SETTINGS_OVERRIDE

public static final Key<Integer> CONTROL_SETTINGS_OVERRIDE

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

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

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

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

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

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

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

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

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

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

This example shows that:

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

Possible values:

• OFF
• ZOOM

Available values for this device:
android.control.availableSettingsOverrides

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

CONTROL_VIDEO_STABILIZATION_MODE

public static final Key<Integer> CONTROL_VIDEO_STABILIZATION_MODE

Whether video stabilization is active.

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

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

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

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

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

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

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

Possible values:

• OFF
• ON
• PREVIEW_STABILIZATION

This key is available on all devices.

CONTROL_ZOOM_RATIO

public static final Key<Float> CONTROL_ZOOM_RATIO

The desired zoom ratio

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

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

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

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

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

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

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

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

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

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

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

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

Range of valid values:
android.control.zoomRatioRange

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

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

DISTORTION_CORRECTION_MODE

public static final Key<Integer> DISTORTION_CORRECTION_MODE

Mode of operation for the lens distortion correction block.

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

OFF means no distortion correction is done.

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

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

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

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

• android.control.afRegions
• android.control.aeRegions
• android.control.awbRegions
• android.scaler.cropRegion
• android.statistics.faces

Possible values:

• OFF
• FAST
• HIGH_QUALITY

Available values for this device:
android.distortionCorrection.availableModes

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

EDGE_MODE

public static final Key<Integer> EDGE_MODE

Operation mode for edge enhancement.

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

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

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

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

Possible values:

• OFF
• FAST
• HIGH_QUALITY
• ZERO_SHUTTER_LAG

Available values for this device:
android.edge.availableEdgeModes

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

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

EXTENSION_CURRENT_TYPE

public static final Key<Integer> EXTENSION_CURRENT_TYPE

Contains the extension type of the currently active extension

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

Range of valid values:
Extension type value listed in CameraExtensionCharacteristics

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

EXTENSION_STRENGTH

public static final Key<Integer> EXTENSION_STRENGTH

Strength of the extension post-processing effect

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

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

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

Range of valid values:
0 - 100

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

FLASH_MODE

public static final Key<Integer> FLASH_MODE

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

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

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

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

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

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

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

Possible values:

• OFF
• SINGLE
• TORCH

This key is available on all devices.

FLASH_STATE

public static final Key<Integer> FLASH_STATE

Current state of the flash unit.

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

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

• Flash-less cameras always return UNAVAILABLE.
• Using android.control.aeMode == ON_ALWAYS_FLASH will always return FIRED.
• Using android.flash.mode == TORCH will always return FIRED.

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

Possible values:

• UNAVAILABLE
• CHARGING
• READY
• FIRED
• PARTIAL

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

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

FLASH_STRENGTH_LEVEL

public static final Key<Integer> FLASH_STRENGTH_LEVEL

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

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

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

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

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

This key is available on all devices.

HOT_PIXEL_MODE

public static final Key<Integer> HOT_PIXEL_MODE

Operational mode for hot pixel correction.

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

Possible values:

• OFF
• FAST
• HIGH_QUALITY

Available values for this device:
android.hotPixel.availableHotPixelModes

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

JPEG_GPS_LOCATION

public static final Key<Location> JPEG_GPS_LOCATION

A location object to use when generating image GPS metadata.

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

This tag is also used for HEIC image capture.

This key is available on all devices.

JPEG_ORIENTATION

public static final Key<Integer> JPEG_ORIENTATION

The orientation for a JPEG image.

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

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

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

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

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

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

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

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

     return jpegOrientation;
 }
 

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

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

Units: Degrees in multiples of 90

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

This key is available on all devices.

JPEG_QUALITY

public static final Key<Byte> JPEG_QUALITY

Compression quality of the final JPEG image.

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

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

This key is available on all devices.

JPEG_THUMBNAIL_QUALITY

public static final Key<Byte> JPEG_THUMBNAIL_QUALITY

Compression quality of JPEG thumbnail.

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

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

This key is available on all devices.

JPEG_THUMBNAIL_SIZE

public static final Key<Size> JPEG_THUMBNAIL_SIZE

Resolution of embedded JPEG thumbnail.

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

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

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

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

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

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

Range of valid values:
android.jpeg.availableThumbnailSizes

This key is available on all devices.

LENS_APERTURE

public static final Key<Float> LENS_APERTURE

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

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

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

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

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

Units: The f-number (f/N)

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

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

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

LENS_DISTORTION

public static final Key<float[]> LENS_DISTORTION

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

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

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

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

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

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

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

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

The distortion model used is the Brown-Conrady model.

Units: Unitless coefficients.

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

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

LENS_FILTER_DENSITY

public static final Key<Float> LENS_FILTER_DENSITY

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

This control will not be supported on most camera devices.

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

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

Units: Exposure Value (EV)

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

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

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

LENS_FOCAL_LENGTH

public static final Key<Float> LENS_FOCAL_LENGTH

The desired lens focal length; used for optical zoom.

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

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

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

Units: Millimeters

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

This key is available on all devices.

LENS_FOCUS_DISTANCE

public static final Key<Float> LENS_FOCUS_DISTANCE

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

Should be zero for fixed-focus cameras

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

Range of valid values:
>= 0

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

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

LENS_FOCUS_RANGE

public static final Key<Pair<Float, Float>> LENS_FOCUS_RANGE

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

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

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

Range of valid values:
>=0

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

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

LENS_INTRINSIC_CALIBRATION

public static final Key<float[]> LENS_INTRINSIC_CALIBRATION

The parameters for this camera device's intrinsic calibration.

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

[f_x, f_y, c_x, c_y, s]
 

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

These are typically used within a transformation matrix K:

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

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

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

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

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

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

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

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

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

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

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

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

LENS_OPTICAL_STABILIZATION_MODE

public static final Key<Integer> LENS_OPTICAL_STABILIZATION_MODE

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

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

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

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

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

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

Possible values:

• OFF
• ON

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

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

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

LENS_POSE_ROTATION

public static final Key<float[]> LENS_POSE_ROTATION

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

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

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

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

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

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

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

p' = Rp

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

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

Units: Quaternion coefficients

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

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

LENS_POSE_TRANSLATION

public static final Key<float[]> LENS_POSE_TRANSLATION

Position of the camera optical center.

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

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

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

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