Research map / Detection and segmentation / 3D detection and driving
Human Correspondence Consensus for 3D Object Semantic Understanding
Yujing Lou, Yang You, Chengkun Li, Zhoujun Cheng, Liangwei Li, Lizhuang Ma, Weiming Wang, Cewu Lu
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 |
|---|---|---|---|
| train with human correspondence sets instead of part labels (ablation) | PointNet · CPNet dense embedding (16 categories subset) | mean Geodesic Error: 0.201 -> 0.099 | Improved |
| use deep networks with geodesic consistency loss instead of SHOT descriptor | DGCNN · CPNet dense embedding | mean Geodesic Error: 0.407 -> 0.099 | Improved |
| use correspondence embeddings instead of part-label embeddings for registration | RANSAC+ICP · Cross-object registration, Chair | relative rotation error (degrees): 20.1 -> 14.6 | Improved |
| use correspondence embeddings instead of part-label embeddings for registration | RANSAC+ICP · Cross-object registration, Airplane | relative rotation error (degrees): 24.4 -> 37.0 | Hurt |
| use correspondence embeddings instead of part-label embeddings for registration | RANSAC+ICP · Cross-object registration, Mug | relative rotation error (degrees): 80.6 -> 17.1 | Improved |
| use correspondence embeddings instead of FPFH | RANSAC+ICP · Cross-object registration, Pistol | relative rotation error (degrees): 9.1 -> 5.3 | Improved |
Closest papers by meaning
- Jamais Vu: Exposing the Generalization Gap in Supervised Semantic Correspondence (NeurIPS 2025 · similarity 0.90)
- SemAlign3D: Semantic Correspondence between RGB-Images through Aligning 3D Object-Class Representations (CVPR 2025 · similarity 0.88)
- Improving Semantic Correspondence with Viewpoint-Guided Spherical Maps (CVPR 2024 · similarity 0.88)
- DenseMatcher: Learning 3D Semantic Correspondence for Category-Level Manipulation from a Single Demo (ICLR 2025 · similarity 0.88)
- Telling Left from Right: Identifying Geometry-Aware Semantic Correspondence (CVPR 2024 · similarity 0.88)
- Bridging Viewpoint Gaps: Geometric Reasoning Boosts Semantic Correspondence (CVPR 2025 · similarity 0.88)
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.
