Research map / 3D vision / Human pose estimation
Sim-to-Real 6D Object Pose Estimation via Iterative Self-Training for Robotic Bin Picking
Kai Chen, Rui Cao, Stephen James, Yichuan Li, Yun-Hui Liu, Pieter Abbeel, 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 2D appearance metric d_a to pose selection (ablation) | self-training with d_g · ROBI | ADD(-S) | Improved |
| add 3D geometry metric d_g to pose selection (ablation) | self-training with d_a · ROBI | ADD(-S) | Improved |
| use both appearance and geometry metrics for selection (ablation) | self-training · ROBI | ADD(-S) | Improved |
| add self-training to DenseFusion | DenseFusion · ROBI mean | ADD(-S): 46.46 -> 62.12 | Improved |
| add self-training to DenseFusion | DenseFusion · ROBI D-Sub Connector | ADD(-S): 7.92 -> 12.66 | Improved |
| add self-training to DenseFusion | DenseFusion · ROBI Eye Bolt | ADD(-S): 60.07 -> 92.36 | Improved |
| add self-training to DC-Net | DC-Net · ROBI | ADD(-S): +16.73% | Improved |
| add self-training to DC-Net | DC-Net · SIBP | ADD(-S): +14.94% | Improved |
| add self-training to the pose network | DC-Net · robot bin-picking grasp success | success rate: 67.96% -> 87.50% | Improved |
| run more self-training iterations (beyond 5) (ablation) | iterative self-training · ROBI | ADD(-S) | No clear effect |
| add a second and third iteration (ablation) | iterative self-training · ROBI | ADD(-S) | Improved |
| use 40% instead of 100% of real data (ablation) | self-training · ROBI (three objects) | ADD(-S) | No clear effect |
Closest papers by meaning
- 6D Object Pose Tracking in Internet Videos for Robotic Manipulation (ICLR 2025 · similarity 0.87)
- GenCOPE: Syn2Real Generalized Category-Level Object Pose Estimation for Robotic Picking (NeurIPS 2026 · similarity 0.86)
- Vision Foundation Model Enables Generalizable Object Pose Estimation (NeurIPS 2024 · similarity 0.86)
- Generalizing 6-DoF Grasp Detection via Domain Prior Knowledge (CVPR 2024 · similarity 0.85)
- CAP-Net: A Unified Network for 6D Pose and Size Estimation of Categorical Articulated Parts from a Single RGB-D Image (CVPR 2025 · similarity 0.85)
- UA-Pose: Uncertainty-Aware 6D Object Pose Estimation and Online Object Completion with Partial References (CVPR 2025 · similarity 0.85)
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.
