Research map / 3D vision / Depth and stereo
Capturing Dynamic Textured Surfaces of Moving Targets
Ruizhe Wang, Lingyu Wei, Etienne Vouga, Qixing Huang, Duygu Ceylan, Gérard Medioni, Hao Li
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 visibility-error-metric PSO registration instead of 4PCS | global registration · Stanford 3D Scanning Repository | success rate (%): 73.0 -> 93.6 | Improved |
| use visibility-error-metric PSO registration instead of 4PCS | global registration · Princeton Shape Benchmark | success rate (%): 73.2 -> 81.5 | Improved |
| use visibility-error-metric PSO registration instead of 4PCS | global registration · Stanford 3D Scanning Repository | average runtime (sec): 10 -> 0.5 | Improved |
| use visibility-error-metric PSO registration instead of FPFH | global registration · Stanford 3D Scanning Repository | success rate (%): 49.3 -> 93.6 | Improved |
| use visibility-error-metric PSO registration instead of Go-ICP | global registration · Stanford 3D Scanning Repository | success rate (%): 34.1 -> 93.6 | Improved |
| use visibility-error-metric PSO registration instead of PCA axes alignment | global registration · Princeton Shape Benchmark | success rate (%): 18.5 -> 81.5 | Improved |
| add guide particles (Levenberg-Marquardt) to traditional PSO (ablation) | modified PSO registration · Stanford 3D Scanning Repository | success rate | Improved |
| add swarm displacement of regular particles on top of guide-particle updates (ablation) | guide-particle-only registration · Stanford 3D Scanning Repository (overlap below 30%) | success rate | Improved |
| add PSO optimization on top of picking the minimum-VEM initial particle (ablation) | registration with initial particles only · Stanford 3D Scanning Repository | success rate | Improved |
| reduce overlap between scans below 40% | previous global registration methods · Stanford and Princeton | success rate | Hurt |
| reduce overlap between scans below 15% | visibility-error-metric registration · Stanford and Princeton | success rate | Hurt |
| use dense-correspondence texture transfer instead of direct texture stitching | texture reconstruction · captured sequences | texture quality | Improved |
Closest papers by meaning
- TexturePose: Supervising Human Mesh Estimation With Texture Consistency (ICCV 2019 · similarity 0.87)
- Dynamic FAUST: Registering Human Bodies in Motion (CVPR 2017 · similarity 0.85)
- REACTO: Reconstructing Articulated Objects from a Single Video (CVPR 2024 · similarity 0.84)
- S3O: A Dual-Phase Approach for Reconstructing Dynamic Shape and Skeleton of Articulated Objects from Single Monocular Video (ICML 2024 · similarity 0.84)
- Link to the Past: Temporal Propagation for Fast 3D Human Reconstruction from Monocular Video (CVPR 2025 · similarity 0.84)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.84)
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.
