PointWAM: 3D 포인트 클라우드 기반 정밀 로봇 조작 모델
PointWAM: 3D World Action Modeling for Dexterous Robotic Manipulation
- 매체
- Hugging Face
- 게시
- 중요도
- 3/5
- 화제 지수
- 10
한국어 요약
PointWAM은 환경과 손을 3D 포인트 궤적으로 분해하여 공동으로 예측하는 3D 월드 액션 모델이다. 인간 시연 영상으로 사전학습하면 DexJoCo 작업에서 기존 대비 11.7점 향상되며, 실제 로봇에서도 강력한 VLA를 능가한다.
원문 발췌 · English
World action models jointly learn to forecast world dynamics and predict robot actions, such that the learned internal world dynamics guide accurate actions. Existing approaches typically represent the world as RGB frames or latent counterparts while predicting actions as end-effector poses or joint angles, but they often struggle to capture the 3D spatial structure and contact geometry central to dexterous manipulation. We introduce Point World Action Model (PointWAM), a 3D world action model that decomposes the world into a scene (i.e., environment) and hands (i.e., actor), and jointly forecasts both as 3D point trajectories within a shared space-time coordinate frame. This explicit, disentangled representation enables effective pre-training on large-scale human demonstration videos without requiring any task-specific object or keypoint selection. Given a colored point cloud and a language instruction, PointWAM predicts how the scene and hands co-evolve in 3D space over time, then retargets the forecast hand motion to robot actions. Pre-training on human videos improves average DexJoCo success by 56.9 percentage points, and scene-trajectory supervision adds 10.9 points over forecasting the hands alone. With both, PointWAM surpasses the prior state of the art on ten DexJoCo tasks by 11.7 points and outperforms strong VLAs on a real robot.