processFile method

@override Future<SegmentationResult> processFile(String filePath, Uint8List frameData, SegmenterInputMetadata metadata)

Process an image file and return segmentation result.

This is more reliable than processFrame as ML Kit can decode the image directly from the file path.

Parameters:

• filePath: Path to image file (JPEG, PNG, etc.)
• frameData: Original RGBA frame data for compositing
• metadata: Image dimensions for mask scaling

Implementation

@override
Future<SegmentationResult> processFile(
  String filePath,
  Uint8List frameData,
  SegmenterInputMetadata metadata,
) async {
  // On iOS/macOS with native initialized, use real segmentation with the file
  if (_supportsNativeSegmentation && _nativeInitialized) {
    try {
      // Read file as bytes and send to native
      final file = File(filePath);
      if (await file.exists()) {
        final imageBytes = await file.readAsBytes();
        return _processWithNativeBytes(imageBytes, frameData, metadata);
      }
    } catch (e) {
      // Fall through to stub
    }
  }

  // Fallback: Return all-white mask
  return _createStubResult(frameData, metadata);
}