Research map / Point clouds / Point cloud classification
ConDor: Self-Supervised Canonicalization of 3D Pose for Partial Shapes
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 |
|---|---|---|---|
| Use ConDor instead of PCA for full-shape canonicalization | ShapeNet full shapes · ShapeNet (13 categories) | IC: 0.0535 -> 0.0308 | Helped |
| Use ConDor instead of PCA for partial-shape canonicalization | ShapeNet partial shapes · ShapeNet (13 categories) | IC: 0.1088 -> 0.0611 | Helped |
| Train the canonicalization system using 50% partiality (ablation) | ConDor · ShapeNet car, partial shapes | average canonicalization error: 0.0594 -> 0.0580 | Helped |
| Add the separation loss (ablation) | ConDor (F+P) · ShapeNet planes, tables, and chairs | average canonicalization error: 0.0700 -> 0.0696 | Helped |
| Add the restriction loss (ablation) | ConDor (F+P) · ShapeNet planes, tables, and chairs | average canonicalization error: 0.0743 -> 0.0696 | Helped |
| Add the separation loss (ablation) | ConDor (F+P) · ShapeNet chair | average canonicalization error: 0.0712 -> 0.0760 | Hurt |
| Predict 5 frames rather than 1 (ablation) | ConDor · ShapeNet lamp, full shapes | IC: 0.0686 -> 0.0403 | Helped |
| Predict 5 frames rather than 1 (ablation) | ConDor · ShapeNet table, full shapes | IC: 0.0607 -> 0.0361 | Helped |
| Train on both full and partial shapes rather than full shapes only (ablation) | ConDor · ShapeNet full shapes | IC: 0.0308 -> 0.0314 | Mixed |
| Use ConDor rather than PCA for classification after canonicalization | ShapeNet (13 categories) · ShapeNet (13 categories) | classification accuracy: 64.9% -> 74.6% | Helped |
Closest papers by meaning
- Canonical Fields: Self-Supervised Learning of Pose-Canonicalized Neural Fields (CVPR 2023 · similarity 0.91)
- One-shot 3D Object Canonicalization based on Geometric and Semantic Consistency (CVPR 2025 · similarity 0.88)
- ShapeMatcher: Self-Supervised Joint Shape Canonicalization Segmentation Retrieval and Deformation (CVPR 2024 · similarity 0.87)
- Learning SO(3)-Invariant Semantic Correspondence via Local Shape Transform (CVPR 2024 · similarity 0.87)
- Emergence of a Shared Canonical Object Frame from In-the-Wild Videos (NeurIPS 2026 · similarity 0.86)
- SIEVE: Overcoming Topological Obstruction in Equivariant Self-Supervised Learning (NeurIPS 2026 · 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.