apple_vision_pose_3d 0.0.1 copy "apple_vision_pose_3d: ^0.0.1" to clipboard
apple_vision_pose_3d: ^0.0.1 copied to clipboard

PlatformiOSmacOS

A Flutter plugin to use Apple Vision Pose Detection to detect the pose of a subject's body in real time from a continuous video or static image.

apple_vision_pose_3d #

Pub Version analysis Star on Github License: MIT

Apple Vision Pose Detection is a Flutter plugin that enables Flutter apps to use Apple Vision Pose Detection 3D.

  • 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: 14.0
  • Xcode 13 or newer
  • Swift 5
  • ML Kit only supports 64-bit architectures (x86_64 and arm64).

iOS

  • Minimum ios Deployment Target: 17.0
  • 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");
  late AppleVisionPose3DController visionController = AppleVisionPose3DController();
  InsertCamera camera = InsertCamera();
  Size imageSize = const Size(640,640*9/16);
  String? deviceId;
  bool loading = true;

  List<PoseData3D>? poseData;
  late double deviceWidth;
  late double deviceHeight;

  @override
  void initState() {
    camera.setupCameras().then((value){
      setState(() {
        loading = false;
      });
      camera.startLiveFeed((InputImage i){
        if(i.metadata?.size != null){
          imageSize = i.metadata!.size;
        }
        if(mounted) {
          Uint8List? image = i.bytes;
          visionController.processImage(image!, i.metadata!.size).then((data){
            poseData = 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(poseData == null || poseData!.isEmpty) return[];
    Map<Joint3D,Color> colors = {
      Joint3D.rightAnkle: Colors.orange,
      Joint3D.rightKnee: Colors.orange,
      Joint3D.rightHip: Colors.orange,

      Joint3D.rightWrist: Colors.purple,
      Joint3D.rightElbow: Colors.purple,

      Joint3D.rightShoulder: Colors.pink,
      Joint3D.leftShoulder: Colors.pink,

      Joint3D.leftElbow: Colors.indigo,
      Joint3D.leftWrist: Colors.indigo,

      Joint3D.leftHip: Colors.grey,
      Joint3D.leftKnee: Colors.grey,
      Joint3D.leftAnkle: Colors.grey,

      Joint3D.root: Colors.yellow,
      Joint3D.centerShoulder: Colors.yellow,
      Joint3D.spine: Colors.yellow,

      Joint3D.centerHead: Colors.cyanAccent,
      Joint3D.topHead: Colors.cyanAccent
    };
    List<Widget> widgets = [];
    for(int j = 0; j < poseData!.length;j++){
      for(int i = 0; i < poseData![j].poses.length; i++){
        widgets.add(
          Positioned(
            top: imageSize.height-poseData![j].poses[i].location.y * imageSize.height,
            left: poseData![j].poses[i].location.x * imageSize.width,
            child: Container(
              width: 10,
              height: 10,
              decoration: BoxDecoration(
                color: colors[poseData![j].poses[i].joint],
                borderRadius: BorderRadius.circular(5)
              ),
            )
          )
        );
      }
    }
    return widgets;
  }

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.

1
likes
150
pub points
19%
popularity

Publisher

unverified uploader

A Flutter plugin to use Apple Vision Pose Detection to detect the pose of a subject's body in real time from a continuous video or static image.

Repository (GitHub)
View/report issues

Documentation

API reference

License

MIT (license)

Dependencies

apple_vision_commons, flutter

More

Packages that depend on apple_vision_pose_3d