PoseDetectorOptions.Builder

AI-generated Key Takeaways

• PoseDetectorOptions.Builder helps configure settings for pose detection, like choosing between processing single images or a stream of related images (e.g., from a camera).

• It allows customization of the detector mode for handling either single images or continuous streams, influencing how input is processed.

• Developers can set preferred hardware configurations for pose detection, prioritizing speed or accuracy based on device capabilities, with options like CPU, GPU, or a combination of both.

• It offers flexibility in task execution by enabling the use of a custom Executor or relying on an internal background thread pool for processing pose detection tasks.

• Using the build() method on the builder, developers can create a PoseDetectorOptions instance with the desired settings to initialize the pose detector.

public static class PoseDetectorOptions.Builder extends Object

Builder for PoseDetectorOptions.

Public Constructor Summary

Builder()

Public Method Summary

PoseDetectorOptions	build()
PoseDetectorOptions.Builder	setDetectorMode(int detectorMode) Set the detector mode to be used, depending on what kind of input is being processed.
PoseDetectorOptions.Builder	setExecutor(Executor executor) Sets the custom Executor to use.
PoseDetectorOptions.Builder	setPreferredHardwareConfigs(int config, int... moreConfigs) Sets preferred PoseDetectorOptionsBase.HardwareConfigs.

Inherited Method Summary

From class java.lang.Object

Object	clone()
boolean	equals(Object arg0)
void	finalize()
final Class<?>	getClass()
int	hashCode()
final void	notify()
final void	notifyAll()
String	toString()
final void	wait(long arg0, int arg1)
final void	wait(long arg0)
final void	wait()

Public Constructors

public Builder ()

Public Methods

public PoseDetectorOptions build ()

public PoseDetectorOptions.Builder setDetectorMode (int detectorMode)

Set the detector mode to be used, depending on what kind of input is being processed. The PoseDetector.process(InputImage) and PoseDetector.process(MlImage) call will treat the inputs differently based on this mode.

Default: PoseDetectorOptionsBase.STREAM_MODE

Use PoseDetectorOptionsBase.STREAM_MODE for input frames consisting of related images, for example, frames from a user's camera. This mode uses tracking, which provides a speed-up when the frames consists of related images without much of a scene change.

Use PoseDetectorOptionsBase.SINGLE_IMAGE_MODE for a single still image, or when the input frames are unrelated.

Parameters

detectorMode	the detector mode used, depending on what kind of input is being processed. One of PoseDetectorOptionsBase.STREAM_MODE or PoseDetectorOptionsBase.SINGLE_IMAGE_MODE.

public PoseDetectorOptions.Builder setExecutor (Executor executor)

Sets the custom Executor to use. If no Executor is set, an internal background thread pool will be used.

Most clients should not need to call this method.

public PoseDetectorOptions.Builder setPreferredHardwareConfigs (int config, int... moreConfigs)

Sets preferred PoseDetectorOptionsBase.HardwareConfigs.

Available options: PoseDetectorOptionsBase.CPU, PoseDetectorOptionsBase.CPU_GPU.

ML Kit will take availability, stability, correctness and latency of each config into consideration and pick the best one from the preferred configs. If none of the preferred configs is applicable, the default CPU config will be used automatically as fallback.

Example usages:

• To let ML Kit pick the best config, no need to call this API.
• If don't want to enable any acceleration, pass in only PoseDetectorOptionsBase.CPU.
• If want to use GPU to offload CPU, even if GPU could be slower, pass in only PoseDetectorOptionsBase.CPU_GPU.