detectFacesWithSegmentationFromCameraFrame method

Future<DetectionWithSegmentationResult> detectFacesWithSegmentationFromCameraFrame(CameraFrame frame, { FaceDetectionMode mode = FaceDetectionMode.full, IsolateOutputFormat outputFormat = IsolateOutputFormat.float32, double binaryThreshold = 0.5, int? maxDim, })

Detects faces and generates a segmentation mask in parallel from a CameraFrame, with all OpenCV work off the UI thread.

The frame's bytes are transferred to both the detection and segmentation isolates; each does its own cvtColor + optional rotate before inference. Total wall time is still ~max(detect, segment) since the two run in parallel.

Implementation

Future<DetectionWithSegmentationResult>
detectFacesWithSegmentationFromCameraFrame(
  CameraFrame frame, {
  FaceDetectionMode mode = FaceDetectionMode.full,
  IsolateOutputFormat outputFormat = IsolateOutputFormat.float32,
  double binaryThreshold = 0.5,
  int? maxDim,
}) {
  _requireReady();
  _requireSegmentationReady();
  return _runTrackedCombinedDetection(
    () => _detectAndSegmentImpl(
      detectOp: 'detectCameraFrame',
      detectFields: _cameraFrameFields(frame, {
        'mode': mode.name,
        'maxDim': maxDim,
      }),
      segmentOp: 'segmentCameraFrame',
      segmentFields: _cameraFrameFields(frame, {
        'outputFormat': outputFormat.index,
        'binaryThreshold': binaryThreshold,
        'maxDim': maxDim,
      }),
    ),
  );
}