hands-3d-pose
High-quality 3D hand pose estimation for egocentric videos from ECCV 2024 (ap229997/hands). Provides 3D joint keypoints and skeleton visualization projected to 2D. Optimized for daily egocentric activities with state-of-the-art accuracy. Outputs hand skeleton overlays on video frames.
npx skills add BioTender-max/awesome-bio-agent-skills --skill hands-3d-pose --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# 3D Hand Pose Estimation (ECCV 2024) ## Overview State-of-the-art 3D hand pose estimation system specifically designed for egocentric (first-person) videos. Published at ECCV 2024, this method provides accurate 3D joint keypoints for hands in daily activities, with robust performance on challenging egocentric viewpoints. The system outputs detailed hand skeleton visualizations with 3D joints projected onto 2D video frames. **Project video**: https://youtu.be/YolFnTtq38E **Key advantage**: Delivers precise joint-level hand pose (not just bounding boxes) for detailed hand motion analysis and gesture understanding. ## When to Use This Skill This skill should be used when: - Need detailed 3D hand joint positions and orientations - Analyzing hand gestures and finger movements in egocentric videos - Building gesture recognition systems with pose-based features - Studying hand-object interactions with precise hand geometry - Creating annotated videos with hand skeleton overlays - Research in egocentric activity recognition - Applications requiring finger-level accuracy (dexterous manipulation) - Biomechanics analysis of hand movements - Sign language or communication gesture analysis **C
- Overview
- When to Use This Skill
- Core Capabilities
- 1. 3D Hand Joint Estimation
- 2. 2D Projection and Visualization
- 3. Video Processing Pipeline
- 4. Single Frame Processing
- 5. Hand Detection Integration
- Installation and Setup
- Usage Examples
- Example 1: Process Video with 3D Pose Output
- Example 2: Extract Hand Pose Features for Gesture Recognition
- Example 3: Analyze Hand-Object Interaction
- Example 4: Batch Process Dataset
Clone repository git clone https://github.com/ap229997/hands.git cd hands Switch to demo branch git checkout demo Install dependencies pip install -r requirements.txt Key dependencies: PyTorch, OpenCV, torchvision, numpy Download pre-trained models bash scripts/download_models.sh
What does the hands-3d-pose skill do?
High-quality 3D hand pose estimation for egocentric videos from ECCV 2024 (ap229997/hands). Provides 3D joint keypoints and skeleton visualization projected to 2D. Optimized for daily egocentric activities with state-of-the-art accuracy. Outputs hand skeleton overlays on video frames.
How do I install it?
Run `npx skills add BioTender-max/awesome-bio-agent-skills --skill hands-3d-pose --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From BioTender-max/awesome-bio-agent-skills, a repository with 135 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.