Efficient Camera Pose Augmentation for View Generalization in Robotic Policy Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910088182104064 |
|---|---|
| author | Wang, Sen Dong, Huaiyi Tian, Jingyi Li, Jiayi Yang, Zhuo Cao, Tongtong Chen, Anlin Wu, Shuang Wang, Le Zhou, Sanping |
| author_facet | Wang, Sen Dong, Huaiyi Tian, Jingyi Li, Jiayi Yang, Zhuo Cao, Tongtong Chen, Anlin Wu, Shuang Wang, Le Zhou, Sanping |
| contents | Prevailing 2D-centric visuomotor policies exhibit a pronounced deficiency in novel view generalization, as their reliance on static observations hinders consistent action mapping across unseen views. In response, we introduce GenSplat, a feed-forward 3D Gaussian Splatting framework that facilitates view-generalized policy learning through novel view rendering. GenSplat employs a permutation-equivariant architecture to reconstruct high-fidelity 3D scenes from sparse, uncalibrated inputs in a single forward pass. To ensure structural integrity, we design a 3D-prior distillation strategy that regularizes the 3DGS optimization, preventing the geometric collapse typical of purely photometric supervision. By rendering diverse synthetic views from these stable 3D representations, we systematically augment the observational manifold during training. This augmentation forces the policy to ground its decisions in underlying 3D structures, thereby ensuring robust execution under severe spatial perturbations where baselines severely degrade. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_29192 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Efficient Camera Pose Augmentation for View Generalization in Robotic Policy Learning Wang, Sen Dong, Huaiyi Tian, Jingyi Li, Jiayi Yang, Zhuo Cao, Tongtong Chen, Anlin Wu, Shuang Wang, Le Zhou, Sanping Robotics Prevailing 2D-centric visuomotor policies exhibit a pronounced deficiency in novel view generalization, as their reliance on static observations hinders consistent action mapping across unseen views. In response, we introduce GenSplat, a feed-forward 3D Gaussian Splatting framework that facilitates view-generalized policy learning through novel view rendering. GenSplat employs a permutation-equivariant architecture to reconstruct high-fidelity 3D scenes from sparse, uncalibrated inputs in a single forward pass. To ensure structural integrity, we design a 3D-prior distillation strategy that regularizes the 3DGS optimization, preventing the geometric collapse typical of purely photometric supervision. By rendering diverse synthetic views from these stable 3D representations, we systematically augment the observational manifold during training. This augmentation forces the policy to ground its decisions in underlying 3D structures, thereby ensuring robust execution under severe spatial perturbations where baselines severely degrade. |
| title | Efficient Camera Pose Augmentation for View Generalization in Robotic Policy Learning |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.29192 |