apple_vision_face_mesh 0.0.2

A Flutter plugin to use Apple Vision Face Detection to detect faces in an image or video stream, identify key facial features, and get the contours of detected faces.

apple_vision_face_mesh

Apple Vision Face Mesh is a Flutter plugin that enables Flutter apps to use tensor flows Face Mesh.

• This plugin is not sponsor or maintained by Apple. The authors are developers who wanted to make a similar plugin to Google's ml kit for macos.

Requirements

MacOS

• Minimum osx Deployment Target: 11.0
• Xcode 13 or newer
• Swift 5
• ML Kit only supports 64-bit architectures (x86_64 and arm64).

iOS

• Minimum ios Deployment Target: 13.0
• Xcode 13 or newer
• Swift 5
• ML Kit only supports 64-bit architectures (x86_64 and arm64).

Getting Started

You need to first import 'package:apple_vision/apple_vision.dart';

final GlobalKey cameraKey = GlobalKey(debugLabel: "cameraKey");
AppleVisionFaceMeshController visionController = AppleVisionFaceMeshController();
InsertCamera camera = InsertCamera();
String? deviceId;
bool loading = true;
Size imageSize = const Size(640,640*9/16);

List<FaceMesh>? faceData;
late double deviceWidth;
late double deviceHeight;

@override
void initState() {
  camera.setupCameras().then((value){
    setState(() {
      loading = false;
    });
    camera.startLiveFeed((InputImage i)async {
      if(i.metadata?.size != null){
        imageSize = i.metadata!.size;
      }
      if(mounted) {
        Uint8List? image = i.bytes;
        await visionController.processImage(image!, i.metadata!.size).then((data){
          faceData = data;
          setState(() {
            
          });
        });
      }
    });
  });
  super.initState();
}
@override
void dispose() {
  camera.dispose();
  super.dispose();
}

@override
Widget build(BuildContext context) {
  deviceWidth = MediaQuery.of(context).size.width;
  deviceHeight = MediaQuery.of(context).size.height;
  return Stack(
    children:<Widget>[
      SizedBox(
        width: imageSize.width, 
        height: imageSize.height, 
        child: loading?Container():CameraSetup(camera: camera, size: imageSize)
    ),
    ]+showPoints()
  );
}

List<Widget> showPoints(){
  if(faceData == null || faceData!.isEmpty) return[];
  List<Widget> widgets = [];
  
  for(int k = 0; k < faceData!.length;k++){
    List<FacePoint> points = faceData![k].mesh;

    for(int j = 0; j < points.length;j++){
      //print(min.width);
      widgets.add(
        Positioned(
          left: points[j].x*imageSize.aspectRatio+faceData![0].image.origin.x+20,
          bottom: imageSize.height/2+90-points[j].y*imageSize.aspectRatio+faceData![0].image.origin.y,
          child: Container(
            width: 2,
            height: 2,
            decoration: BoxDecoration(
              color: Colors.white,
              borderRadius: BorderRadius.circular(1)
            ),
          )
        )
      );
    }
  }
  return widgets;
}

Widget loadingWidget(){
  return Container(
    width: deviceWidth,
    height: deviceHeight,
    color: Theme.of(context).canvasColor,
    alignment: Alignment.center,
    child: const CircularProgressIndicator(color: Colors.blue)
  );
}

Example

Find the example for this API here.

Contributing

Contributions are welcome. In case of any problems look at existing issues, if you cannot find anything related to your problem then open an issue. Create an issue before opening a pull request for non trivial fixes. In case of trivial fixes open a pull request directly.