Research map / 3D vision / 3D generation
Learning Surgical Robotic Manipulation with 3D Spatial Priors
Yu Sheng, Lidian Wang, Xiaomeng Chu, Jiajun Deng, Min Cheng, Yanyong Zhang, Bei Hua, Houqiang Li, Jianmin Ji
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| finetune the geometry transformer on Surgical3D (ablation) | SST · Peg pickup Test2 | success rate: 0/10 -> 8/10 | Improved |
| finetune the geometry transformer on Surgical3D (ablation) | SST · Knot tying | success rate: grasp 0/10 -> 10/10; loop 0/10 -> 7/10 | Improved |
| finetune the geometry transformer on Surgical3D (ablation) | MASt3R · Surgical3D reconstruction | Accuracy error: 0.0111 -> 0.0048 | Improved |
| use multi-level spatial feature connector instead of last-layer features (ablation) | SST · Peg pickup Test2 | success rate: 0/10 -> 8/10 | Improved |
| use MSFC concatenation instead of multi-layer separate cross-attention (ablation) | SST · Knot tying loop | success rate: 0/10 -> 7/10 | Improved |
| use MASt3R instead of VGGT as geometry transformer (ablation) | SST · Inference latency | latency (ms): 140.4 -> 56.2 | Improved |
| add wrist cameras (SRT) versus SST endoscope only | surgical policy · Peg pickup Test1 | success rate: 10/10 vs 10/10 | No clear effect |
| SST versus ACT and Diffusion Policy trained on endoscope images only | surgical policy · Knot tying and gallbladder dissection | success rate: ACT whole task 0/10 -> SST 6/10 (gallbladder) | Improved |
Closest papers by meaning
- SurgVista: Long-Horizon Surgical World Modeling with Plausible Instrument-Tissue Dynamics (NeurIPS 2026 · similarity 0.85)
- Redefining the Laparoscopic Spatial Sense: AI-Based Intra- and Postoperative Measurement from Stereoimages (AAAI 2024 · similarity 0.85)
- Towards Dynamic 3D Reconstruction of Hand-Instrument Interaction in Ophthalmic Surgery (NeurIPS 2025 · similarity 0.83)
- CholecTrack20: A Multi-Perspective Tracking Dataset for Surgical Tools (CVPR 2025 · similarity 0.83)
- 3D-MVP: 3D Multiview Pretraining for Manipulation (CVPR 2025 · similarity 0.83)
- Lift3D Policy: Lifting 2D Foundation Models for Robust 3D Robotic Manipulation (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
