Research map / Point clouds / Shape and surface completion
Learning Parallel Dense Correspondence From Spatio-Temporal Descriptors for Efficient and Robust 4D Reconstruction
Jiapeng Tang, Dan Xu, Kui Jia, Lei Zhang
Check 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 LPDC-Net instead of OFlow | 4D reconstruction · D-FAUST, time-evenly sampled, S1 | IoU: 81.5% -> 84.9% | Improved |
| use LPDC-Net instead of OFlow | 4D reconstruction · D-FAUST, time-evenly sampled, S1 | Chamfer distance: 0.065 -> 0.055 | Improved |
| use LPDC-Net instead of OFlow | 4D reconstruction · D-FAUST, time-unevenly sampled, S1 | IoU: 76.9% -> 83.8% | Improved |
| use LPDC-Net instead of OFlow | 4D reconstruction · D-FAUST, time-unevenly sampled, S2 | Correspondence distance: 0.178 -> 0.118 | Improved |
| use parallel occupancy-field transformations rather than OFlow's sequential ODE-based motion modeling | 4D reconstruction inference · D-FAUST test sequences | Inference time (s): 1.84 -> 0.23 | Improved |
| add the proposed spatio-temporal encoder (ablation) | LPDC-Net 4D shape completion · D-FAUST, time-unevenly sampled, S1 | IoU: 81.0% -> 82.4% | Improved |
| add the proposed spatio-temporal encoder (ablation) | LPDC-Net 4D shape completion · D-FAUST, time-unevenly sampled, S2 | IoU: 71.7% -> 72.9% | Improved |
| use parallel correspondence modeling rather than sequential adjacent-frame transformations (ablation) | LPDC-Net 4D shape completion · D-FAUST, time-unevenly sampled, S1 | Correspondence distance: 0.254 -> 0.105 | Improved |
| use parallel correspondence modeling rather than sequential adjacent-frame transformations (ablation) | LPDC-Net 4D shape completion · D-FAUST, time-unevenly sampled, S2 | Correspondence distance: 0.344 -> 0.134 | Improved |
Closest papers by meaning
- Occupancy Flow: 4D Reconstruction by Learning Particle Dynamics (ICCV 2019 · similarity 0.90)
- Motion2VecSets: 4D Latent Vector Set Diffusion for Non-rigid Shape Reconstruction and Tracking (CVPR 2024 · similarity 0.88)
- 4D-LRM: Large Space-Time Reconstruction Model From and To Any View at Any Time (NeurIPS 2025 · similarity 0.88)
- IGGT4D: Streaming 4D Instance-Grounded Geometry Transformer (NeurIPS 2026 · similarity 0.87)
- Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos (CVPR 2025 · similarity 0.87)
- Dynamic Neural Surfaces for Elastic 4D Shape Representation and Analysis (CVPR 2025 · similarity 0.87)
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.
