Yang Wang

3D Gaussian Digital Twins for View-Robust 3D Visuomotor Policy Learning on Articulated Objects

Single-view 3D visuomotor policies based on point-cloud observations, such as 3D Diffusion Policy, often suffer from limited viewpoint generalization due to partial observability and occlusion. This is particularly challenging for articulated objects whose internal motion induces geometry changes. To address this probl...