Spaces:
Running
Running
| title: Pose-Think AI Movement Analysis | |
| emoji: 🎯 | |
| colorFrom: blue | |
| colorTo: green | |
| sdk: gradio | |
| sdk_version: 4.8.0 | |
| app_file: app.py | |
| pinned: false | |
| license: mit | |
| # 🎯 Pose-Think: AI-Powered Movement Analysis Suite | |
| Real-time posture and movement analysis with multiple specialized modes. Get instant feedback on what the camera sees! | |
| ## 🌟 Features | |
| ### 🎯 **Basic Posture Analysis** | |
| - **Real-time body detection**: Instantly identifies visible body parts | |
| - **Joint angle measurements**: Precise elbow, knee, and other joint angles | |
| - **Posture alignment**: Shoulder level, hip alignment, neck position | |
| - **Clear feedback**: "Left shoulder high", "Neck tilted right", "Elbows at 156°" | |
| ### 🎯 **Enhanced Posture Analysis** | |
| - **Everything from Basic mode** plus: | |
| - **Age-specific recommendations**: Tailored advice for different age groups | |
| - **BMI calculation**: Height/weight analysis for posture load assessment | |
| - **Personalized insights**: Context-aware health recommendations | |
| ### 🤚 **Hand Tracking** | |
| - **Dual hand detection**: Track up to 2 hands simultaneously | |
| - **Finger counting**: Automatic detection of extended fingers | |
| - **Gesture analysis**: Real-time hand position and orientation | |
| - **Perfect for**: Hand exercises, gesture recognition, rehabilitation | |
| ## 🚀 How to Use | |
| 1. **Choose Analysis Type**: Select from Basic Posture, Enhanced Posture, or Hand Tracking | |
| 2. **Allow Camera Access**: Grant permission when prompted by your browser | |
| 3. **Position Yourself**: | |
| - **For Posture**: Stand 2-3 meters from camera, full body visible | |
| - **For Hands**: Show hands clearly to the camera | |
| 4. **Click Analyze**: Get instant, detailed feedback | |
| 5. **Enhanced Mode**: Optionally enter age/height/weight for personalized insights | |
| ## 📊 What You'll See | |
| ### Basic/Enhanced Posture Analysis: | |
| ``` | |
| ✅ Visible: Head, Shoulders, Elbows, Hips, Knees | |
| 📐 Left elbow angle: 156.3° | |
| 📐 Right elbow angle: 162.1° | |
| ⚠️ Left shoulder higher | |
| ✅ Hips level | |
| 🔍 Neck: Tilted right | |
| ``` | |
| ### Enhanced Mode Additional Info: | |
| ``` | |
| 👤 Profile: Age: 32 | BMI: 26.1 | |
| ⚠️ BMI slightly high - extra load on posture | |
| 🎯 Age-Specific Recommendations: | |
| 💡 Middle age: Regular exercise important | |
| 💼 Make workspace ergonomic | |
| ``` | |
| ### Hand Tracking: | |
| ``` | |
| ✅ 2 hands detected | |
| 🖐️ Hand 1: 5 fingers up | |
| 🖐️ Hand 2: 3 fingers up | |
| ``` | |
| ## 🎯 Perfect For | |
| - **Personal Health**: Quick posture checks and movement awareness | |
| - **Workplace Wellness**: Ergonomic assessments and posture monitoring | |
| - **Fitness & Rehabilitation**: Exercise form checking and progress tracking | |
| - **Education & Research**: Movement analysis and biomechanics studies | |
| - **Accessibility**: Hand gesture recognition and interaction | |
| ## 🔧 Technical Details | |
| - **AI Model**: MediaPipe Pose and Hands (Google) | |
| - **Processing**: Real-time analysis at 30+ FPS | |
| - **Accuracy**: 95%+ detection rate in good lighting | |
| - **Privacy**: All processing done locally, no data stored | |
| - **Compatibility**: Works on desktop, tablet, and mobile browsers | |
| ## 📝 Tips for Best Results | |
| ### For Posture Analysis: | |
| - **Lighting**: Ensure even, bright lighting | |
| - **Background**: Use plain, contrasting background | |
| - **Distance**: Stand 2-3 meters from camera | |
| - **Position**: Keep full body visible in frame | |
| ### For Hand Tracking: | |
| - **Visibility**: Keep hands clearly visible | |
| - **Contrast**: Ensure hands contrast with background | |
| - **Movement**: Avoid rapid movements for better tracking | |
| - **Lighting**: Good lighting improves accuracy | |
| ## 🎓 Understanding the Feedback | |
| ### Symbols: | |
| - ✅ **Green checkmarks**: Good alignment/posture | |
| - ⚠️ **Warning signs**: Issues detected that need attention | |
| - 📐 **Measurements**: Precise joint angles in degrees | |
| - 🔍 **Position info**: Head, neck, body position details | |
| - 👤 **Profile info**: Personal context and recommendations | |
| ### Measurements: | |
| - **Joint angles**: 0-180° range, closer to 180° = straighter | |
| - **Alignment**: Compares left vs right side positioning | |
| - **Visibility**: Shows which body parts are clearly detected | |
| ## 🚀 Built With | |
| - **MediaPipe**: Google's ML framework for pose and hand detection | |
| - **Gradio**: Modern web interface for ML applications | |
| - **OpenCV**: Computer vision and image processing | |
| - **NumPy**: Numerical computations | |
| ## 📄 License | |
| MIT License - feel free to use and modify for your projects! | |
| ## 🤝 Contributing | |
| This is part of the Pose-Think project. Contributions welcome on [GitHub](https://github.com/semskurto/pose-think)! | |
| --- | |
| **⚠️ Disclaimer**: This tool is for educational and wellness purposes only. It does not provide medical diagnosis. Consult healthcare professionals for medical concerns. | |