flutter_pose_detection 0.1.0

Hardware-accelerated pose detection using native ML frameworks (Apple Vision on iOS, TensorFlow Lite on Android). Detects 33 body landmarks in MediaPipe-compatible format.

Flutter Pose Detection

Hardware-accelerated pose detection Flutter plugin using native ML frameworks.

Features

• Hardware Acceleration: Automatic NPU/GPU acceleration on supported devices
• Cross-Platform: iOS (Vision Framework) and Android (TensorFlow Lite)
• 33 Landmarks: MediaPipe-compatible body pose format
• Multiple Modes: Image, camera stream, and video file analysis
• High Performance: 15+ FPS realtime, <50ms single image

Platform Support

Platform	ML Framework	Acceleration
iOS 14+	Vision Framework	Neural Engine (automatic)
Android API 26+	TensorFlow Lite	GPU Delegate / NNAPI / CPU

Installation

dependencies:
  flutter_pose_detection: ^0.1.0

iOS Setup

Add to ios/Podfile:

platform :ios, '14.0'

Add camera permission to ios/Runner/Info.plist:

<key>NSCameraUsageDescription</key>
<string>Camera access is needed for pose detection</string>

Android Setup

Update android/app/build.gradle:

android {
    defaultConfig {
        minSdkVersion 26
    }
}

Add camera permission to android/app/src/main/AndroidManifest.xml:

<uses-permission android:name="android.permission.CAMERA" />

Quick Start

import 'package:flutter_pose_detection/flutter_pose_detection.dart';

// Create and initialize detector
final detector = NpuPoseDetector();
await detector.initialize();

// Detect pose from image
final imageBytes = await File('image.jpg').readAsBytes();
final result = await detector.detectPose(imageBytes);

if (result.hasPoses) {
  final pose = result.firstPose!;
  print('Detected ${pose.landmarks.length} landmarks');

  // Access specific landmarks
  final nose = pose.getLandmark(LandmarkType.nose);
  print('Nose at (${nose.x}, ${nose.y})');
}

// Clean up
detector.dispose();

Documentation

• API Reference
• Quickstart Guide
• Landmark Reference

License

MIT License - see LICENSE