Research map / 3D vision / Depth and stereo
Spatiotemporal Bundle Adjustment for Dynamic 3D Reconstruction
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 a 3D motion prior instead of frame-level geometry triangulation | Spatiotemporal bundle adjustment · CMU Motion Capture synthetic dynamic points | reprojection error: 17.8 -> 0.85 | Helped |
| Use a 3D motion prior instead of cubic B-spline trajectories | Spatiotemporal bundle adjustment · CMU Motion Capture synthetic dynamic points | reprojection error: 1.68 -> 0.85 | Helped |
| Add final joint optimization and DCT trajectory resampling (ablation) | Motion Prior · CMU Motion Capture synthetic dynamic points | reprojection error: 0.85 -> 0.74 | Helped |
| Add final joint optimization and DCT trajectory resampling (ablation) | Motion Prior · Dance | RMSE (pixels), dynamic points: 2.12 -> 1.71 | Helped |
| Add final joint optimization and DCT trajectory resampling (ablation) | Motion Prior · Jump | RMSE (pixels), dynamic points: 1.34 -> 1.26 | Helped |
| Use least kinetic energy rather than least action as the motion prior (ablation) | Temporal alignment of sampled 3D trajectories · CMU Motion Capture | temporal offset error | Helped |
| Use least kinetic energy or least force rather than least action (ablation) | Temporal alignment of sampled 3D trajectories · CMU Motion Capture | temporal offset error | Helped |
| Use motion-prior reconstruction instead of geometry triangulation | Spatiotemporal bundle adjustment · Checkerboard | RMSE (pixels), dynamic points: 6.59 -> 1.21 | Helped |
| Use motion-prior reconstruction instead of geometry triangulation | Spatiotemporal bundle adjustment · Jump | RMSE (pixels), dynamic points: 1.91 -> 1.34 | Helped |
| Use motion-prior reconstruction instead of geometry triangulation | Spatiotemporal bundle adjustment · Dance | RMSE (pixels), dynamic points: 5.23 -> 2.12 | Helped |
| Use motion-prior reconstruction with correct sub-frame alignment instead of incorrect alignment (ablation) | 3D trajectory estimation · Checkerboard sequence | trajectory loops | Helped |
Closest papers by meaning
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.88)
- Enhancing 3D Reconstruction for Dynamic Scenes (NeurIPS 2025 · similarity 0.88)
- Flux4D: Flow-based Unsupervised 4D Reconstruction (NeurIPS 2025 · similarity 0.86)
- ProDyG: Progressive Dynamic Scene Reconstruction via Gaussian Splatting from Monocular Videos (NeurIPS 2025 · similarity 0.86)
- Motion Matters: Compact Gaussian Streaming for Free-Viewpoint Video Reconstruction (NeurIPS 2025 · similarity 0.86)
- Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos (CVPR 2025 · 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.