Research map / 3D vision / Depth and stereo
Iterative Corresponding Geometry: Fusing Region and Depth for Highly Efficient 3D Tracking of Textureless Objects
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Fuse region and depth modalities (ablation) | ICG · OPT | AUC: 8.94 -> 16.54 | Helped |
| Add region modality (ablation) | ICG · YCB-Video | ADD AUC (%): 66.1 -> 86.4 | Helped |
| Add depth modality (ablation) | ICG · Choi | Translation RMS: 41.65 -> 0.04 | Helped |
| Add depth modality (ablation) | ICG · YCB-Video | ADD-S AUC (%): 42.8 -> 96.5 | Helped |
| Add regularization (ablation) | ICG · OPT | AUC: 14.48 -> 16.54 | Helped |
| Add regularization (ablation) | ICG · YCB-Video | ADD AUC (%): 72.0 -> 86.4 | Helped |
| Add regularization (ablation) | ICG · Choi | Translation RMS: 0.04 -> 0.04 | No clear effect |
| Add occlusion handling (ablation) | ICG · YCB-Video | ADD-S AUC (%): 91.9 -> 96.5 | Helped |
| Use ICG instead of SRT3D | 3D tracking · OPT average | AUC: 15.73 -> 16.54 | Helped |
| Use ICG instead of se(3)-TrackNet | YCB-Video tracking · YCB-Video all frames | ADD-S AUC (%): 95.7 -> 96.5 | Helped |
| Run tracking on a single CPU core without a GPU | ICG · YCB-Video | Speed (Hz): 90.9 Hz -> 788.4 Hz | Helped |
Closest papers by meaning
- Prior-free 3D Object Tracking (CVPR 2025 · similarity 0.87)
- iG-6DoF: Model-free 6DoF Pose Estimation for Unseen Object via Iterative 3D Gaussian Splatting (CVPR 2025 · similarity 0.85)
- Robust 3D Tracking with Quality-Aware Shape Completion (AAAI 2024 · similarity 0.85)
- SpatialTracker: Tracking Any 2D Pixels in 3D Space (CVPR 2024 · similarity 0.85)
- TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels (NeurIPS 2025 · similarity 0.84)
- ProxyPose: 6-DoF Pose Tracking via Video-to-Video Translation (NeurIPS 2026 · similarity 0.84)
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.