Research map / Point clouds / Shape and surface completion
Learning 3D Dense Correspondence via Canonical Point Autoencoder
Read the paperCheck 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 |
|---|---|---|---|
| add cross-reconstruction loss L_cross (ablation) | CPAE · KeypointNet chair keypoint transfer | keypoint transfer accuracy | Helped |
| add cross-reconstruction loss under unaligned (rotated) inputs (ablation) | CPAE · KeypointNet chair, rotation noise N(0,0.5^2) | keypoint transfer accuracy | Helped |
| use adaptive alpha (1 decayed to 0) instead of constant alpha=0 (ablation) | CPAE · KeypointNet chair keypoint transfer | keypoint transfer accuracy | Helped |
| use adaptive alpha instead of constant alpha=1 (ablation) | CPAE · KeypointNet chair keypoint transfer | keypoint transfer accuracy | Helped |
| use 3D UV sphere instead of 2D UV grid as canonical primitive (ablation) | CPAE · KeypointNet chair keypoint transfer | keypoint transfer accuracy | Helped |
| CPAE vs Liu et al. part label transfer | CPAE · ShapeNet part label transfer | average IoU (%): 61.7 -> 65.8 | Helped |
| CPAE vs Liu et al. keypoint transfer at threshold 0.05 | CPAE · KeypointNet | keypoint transfer accuracy: 11.2% more accurate on average | Helped |
Closest papers by meaning
- Learning Implicit Functions for Topology-Varying Dense 3D Shape Correspondence (NeurIPS 2020 · similarity 0.90)
- Learning SO(3)-Invariant Semantic Correspondence via Local Shape Transform (CVPR 2024 · similarity 0.89)
- CorrNet3D: Unsupervised End-to-End Learning of Dense Correspondence for 3D Point Clouds (CVPR 2021 · similarity 0.89)
- Jamais Vu: Exposing the Generalization Gap in Supervised Semantic Correspondence (NeurIPS 2025 · similarity 0.88)
- Adaptive Siamese Masked Autoencoder with Global Optimization for Unsupervised Point Cloud Shape Correspondence (AAAI 2025 · similarity 0.87)
- Improving Semantic Correspondence with Viewpoint-Guided Spherical Maps (CVPR 2024 · similarity 0.86)
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.