ISS Policy : Scalable Diffusion Policy with Implicit Scene Supervision
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908858076626944 |
|---|---|
| author | Xia, Wenlong Zhang, Jinhao Zhang, Ce Wang, Yaojia Li, Huizhe Gong, Youmin Mei, Jie |
| author_facet | Xia, Wenlong Zhang, Jinhao Zhang, Ce Wang, Yaojia Li, Huizhe Gong, Youmin Mei, Jie |
| contents | Vision-based imitation learning has enabled impressive robotic manipulation skills, but its reliance on object appearance while ignoring the underlying 3D scene structure leads to low training efficiency and poor generalization. To address these challenges, we introduce \emph{Implicit Scene Supervision (ISS) Policy}, a 3D visuomotor DiT-based diffusion policy that predicts sequences of continuous actions from point cloud observations. We extend DiT with a novel implicit scene supervision module that encourages the model to produce outputs consistent with the scene's geometric evolution, thereby improving the performance and robustness of the policy. Notably, ISS Policy achieves state-of-the-art performance on both single-arm manipulation tasks (MetaWorld) and dexterous hand manipulation (Adroit). In real-world experiments, it also demonstrates strong generalization and robustness. Additional ablation studies show that our method scales effectively with both data and parameters. Code and videos will be released. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_15020 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ISS Policy : Scalable Diffusion Policy with Implicit Scene Supervision Xia, Wenlong Zhang, Jinhao Zhang, Ce Wang, Yaojia Li, Huizhe Gong, Youmin Mei, Jie Robotics Vision-based imitation learning has enabled impressive robotic manipulation skills, but its reliance on object appearance while ignoring the underlying 3D scene structure leads to low training efficiency and poor generalization. To address these challenges, we introduce \emph{Implicit Scene Supervision (ISS) Policy}, a 3D visuomotor DiT-based diffusion policy that predicts sequences of continuous actions from point cloud observations. We extend DiT with a novel implicit scene supervision module that encourages the model to produce outputs consistent with the scene's geometric evolution, thereby improving the performance and robustness of the policy. Notably, ISS Policy achieves state-of-the-art performance on both single-arm manipulation tasks (MetaWorld) and dexterous hand manipulation (Adroit). In real-world experiments, it also demonstrates strong generalization and robustness. Additional ablation studies show that our method scales effectively with both data and parameters. Code and videos will be released. |
| title | ISS Policy : Scalable Diffusion Policy with Implicit Scene Supervision |
| topic | Robotics |
| url | https://arxiv.org/abs/2512.15020 |