Research map / 3D vision / 3D generation
V-DPM: 4D Video Reconstruction with Dynamic Point Maps
Edgar Sucar, Eldar Insafutdinov, Zihang Lai, Andrea Vedaldi
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 |
|---|---|---|---|
| process whole 10-frame video jointly instead of independent frame pairs (ablation) | V-DPM · PointOdyssey 10-frame tracking | tracking EPE: 0.037 -> 0.032 | Improved |
| process whole video jointly instead of frame pairs (ablation) | V-DPM · Kubric-F 10-frame tracking | tracking EPE: 0.066 -> 0.027 | Improved |
| process whole video jointly instead of frame pairs (ablation) | V-DPM · Waymo 10-frame tracking | tracking EPE: 0.094 -> 0.042 | Improved |
| fine-tune static VGGT with time-conditioned decoder on dynamic data | VGGT · PointOdyssey 2-view, margin 2 | EPE P0(t0): 0.115 (DPM) -> 0.029 | Improved |
| V-DPM vs DPM on whole video tracking (DPM limited to pairs) | DPM · Kubric-F 10-frame tracking | tracking EPE: 0.088 -> 0.027 | Improved |
| extend VGGT to V-DPM for video depth | VGGT · Sintel video depth | Abs Rel: 0.242 -> 0.247 | Mixed |
| extend VGGT to V-DPM for camera pose | VGGT · Sintel camera pose | ATE: 0.167 -> 0.105 | Mixed |
Later papers that compared with this method
4 later accepted papers report VDPM as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- OmniX: Any-view and Any-time 4D reconstruction via Feed-forward Trajectory Fields (ECCV 2026 · better on 13 of 13 shared results)
- 4RC: 4D Reconstruction via Conditional Querying Anytime and Anywhere (ICML 2026 · better on 5 of 7 shared results)
- Face Anything: 4D Face Reconstruction from Any Image Sequence (ECCV 2026 · better on 7 of 7 shared results)
- Track4World: Feedforward World-centric Dense 3D Tracking of All Pixels (ECCV 2026 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 2025) · better on 1 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Enhancing 3D Reconstruction for Dynamic Scenes (NeurIPS 2025 · similarity 0.90)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.90)
- DualPM: Dual Posed-Canonical Point Maps for 3D Shape and Pose Reconstruction (CVPR 2025 · similarity 0.89)
- MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion (ICLR 2025 · similarity 0.89)
- Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos (CVPR 2025 · similarity 0.89)
- Flux4D: Flow-based Unsupervised 4D Reconstruction (NeurIPS 2025 · similarity 0.88)
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.
