Research map / Point clouds / Shape and surface completion
CaSPR: Learning Canonical Spatiotemporal Point Cloud Representations
Davis Rempe, Tolga Birdal, Yongheng Zhao, Zan Gojcic, Srinath Sridhar, Leonidas Guibas
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 |
|---|---|---|---|
| Use TPointNet++ instead of PointNet++ w/ Time | T-NOCS canonicalization · Cars | Spatial Err: 0.0510 -> 0.0118 | Improved |
| Use CaSPR instead of PointFlow | Sequence reconstruction · Cars, 10 observed frames | EMD: 12.838 -> 10.103 | Improved |
| Use CaSPR instead of PointFlow | Sequence reconstruction · Chairs, 10 observed frames | EMD: 17.267 -> 13.009 | Improved |
| Use CaSPR instead of PointFlow | Sequence reconstruction · Airplanes, 10 observed frames | EMD: 9.500 -> 6.026 | Improved |
| Use CaSPR rather than OFlow | Warping Cars reconstruction · 10 observed timesteps | CD: 1.512 -> 0.955 | Improved |
| Reverse the input sequence timestamps | CaSPR · Held-out rigid car sequences | EMD: 7.114 -> 88.938 | Hurt |
| Add canonicalization loss (ablation) | CaSPR · Rigid cars, 10 observed frames | CD: 0.605 -> 0.566 | Improved |
| Add canonicalization loss (ablation) | CaSPR · Rigid cars, 10 observed frames | EMD: 11.482 -> 10.103 | Improved |
| Add static/dynamic latent factorization (ablation) | CaSPR · Rigid cars, 10 observed frames | CD: 0.635 -> 0.566 | Improved |
| Add static/dynamic latent factorization (ablation) | CaSPR · Rigid cars, 10 observed frames | NFE: 101.3 -> 39.6 | Improved |
| Add pairwise input augmentation (ablation) | CaSPR · Rigid cars, 10 observed frames | CD: 0.577 -> 0.566 | Improved |
| Add pairwise input augmentation (ablation) | CaSPR · Rigid cars, 10 observed frames | EMD: 10.253 -> 10.103 | Improved |
Closest papers by meaning
- Motion2VecSets: 4D Latent Vector Set Diffusion for Non-rigid Shape Reconstruction and Tracking (CVPR 2024 · similarity 0.86)
- Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis (CVPR 2025 · similarity 0.85)
- Living Scenes: Multi-object Relocalization and Reconstruction in Changing 3D Environments (CVPR 2024 · similarity 0.85)
- Adaptive 3D Reconstruction via Diffusion Priors and Forward Curvature-Matching Likelihood Updates (NeurIPS 2025 · similarity 0.85)
- Rectified Point Flow: Generic Point Cloud Pose Estimation (NeurIPS 2025 · similarity 0.85)
- Learning Shape-Independent Transformation via Spherical Representations for Category-Level Object Pose Estimation (ICLR 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.
