Research map / 3D vision / Human pose estimation
Shape-guided Configuration-aware Learning for Endoscopic-image-based Pose Estimation of Flexible Robotic Instruments
Yiyao Ma, Kai Chen, Hon-Sing Tong, Ruofeng Wei, Yui-Lun Ng, Ka-Wai Kwok, Qi Dou
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 |
|---|---|---|---|
| add configuration-aware 3D shape guidance to image features (ablation) | flexible robot pose regression · Endoscopic flexible robot dataset | median error / accuracy | Improved |
| add robot configuration information to aggregate part-level features in shape guidance (ablation) | shape guidance · Endoscopic flexible robot dataset | accuracy | Improved |
| use prior-based shape guidance instead of lifting to 3D with a predicted depth map (ablation) | PoseEst. and PoseRefine. · Endoscopic flexible robot dataset | accuracy | Improved |
| add shape-deformation pose refinement stage (ablation) | PoseEst. · Endoscopic flexible robot G1 (delta) | mean angular error (deg): 5.86 -> 4.08 | Improved |
| add shape-deformation pose refinement stage (ablation) | PoseEst. · Endoscopic flexible robot G2 (gamma) | median angular error (deg): 2.00 -> 2.37 | Hurt |
| use regression with shape prior instead of keypoint detection or skeleton extraction | flexible robot pose estimation · Endoscopic flexible robot G1 (delta) | mean angular error (deg): 49.34 -> 5.86 | Improved |
| use PoseEst. instead of SimPS under occlusion | flexible robot pose estimation · Occlusion (flushing water and bubbles) | mean angular error (deg): 23.26 -> 18.68 | Improved |
Closest papers by meaning
- Learning Surgical Robotic Manipulation with 3D Spatial Priors (CVPR 2026 · similarity 0.86)
- iMED: A Multi-Endoscope Dataset for Surgical 3D Perception (ECCV 2026 · similarity 0.83)
- RoboPEPP: Vision-Based Robot Pose and Joint Angle Estimation through Embedding Predictive Pre-Training (CVPR 2025 · similarity 0.82)
- SurgVista: Long-Horizon Surgical World Modeling with Plausible Instrument-Tissue Dynamics (NeurIPS 2026 · similarity 0.82)
- ESCAPE: Equivariant Shape Completion via Anchor Point Encoding (CVPR 2025 · similarity 0.81)
- Redefining the Laparoscopic Spatial Sense: AI-Based Intra- and Postoperative Measurement from Stereoimages (AAAI 2024 · similarity 0.80)
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.
