Research map / 3D vision / Gaussian splatting
Efficiently Reconstructing Dynamic Scenes One D4RT at a Time
Chuhan Zhang, Guillaume Le Moing, Skanda Koppula, Ignacio Rocco, Liliane Momeni, Junyu Xie, Shuyang Sun, Rahul Sukthankar, Joëlle K. Barral, Raia Hadsell, Zoubin Ghahramani, Andrew Zisserman and 2 more
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 |
|---|---|---|---|
| add local RGB appearance patch to decoder queries (ablation) | D4RT ViT-L · Sintel video depth | AbsRel (S): 0.366 -> 0.302 | Improved |
| add local RGB appearance patch to decoder queries (ablation) | D4RT ViT-L · Sintel camera pose | ATE: 0.173 -> 0.091 | Improved |
| initialize encoder from VideoMAE instead of random weights (ablation) | D4RT · Sintel video depth | AbsRel (S): 0.738 -> 0.302 | Improved |
| initialize encoder from VideoMAE instead of random weights (ablation) | D4RT · Sintel camera pose | RPE-R: 1.126 -> 0.245 | Improved |
| scale encoder from ViT-B to ViT-g (ablation) | D4RT · Sintel video depth | AbsRel (S): 0.319 -> 0.191 | Improved |
| add 2D position auxiliary loss (ablation) | D4RT · Sintel video depth | AbsRel (S): +0.071 when removed | Improved |
| add normal auxiliary loss (ablation) | D4RT · Sintel video depth | AbsRel (S): +0.043 when removed | Improved |
| add confidence auxiliary loss (ablation) | D4RT · Sintel camera pose | ATE: +0.126 when removed | Improved |
| add confidence auxiliary loss (ablation) | D4RT · Sintel video depth | AbsRel (SS): -0.025 when removed | Hurt |
| add visibility auxiliary loss (ablation) | D4RT · Sintel camera pose and depth | RPE-R / AbsRel (SS): +0.022 RPE-R, -0.025 AbsRel (SS) when removed | Mixed |
| add displacement auxiliary loss (ablation) | D4RT · Sintel video depth and pose | AbsRel (S) / ATE: +0.011 AbsRel (S), +0.011 ATE when removed | Improved |
| add RGB patches at original resolution at decoding (ablation) | D4RT ViT-g · Sintel depth boundary | PDBE (scale): 2.254 -> 2.193 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- SpatialTrackerV2 — SpatialTracker: Tracking Any 2D Pixels in 3D Space (CVPR 2024) · better on 6 of 6 shared results
- MegaSaM — MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos (CVPR 2025) · better on 4 of 4 shared results
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 2025) · better on 4 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
- 4DGT: Learning a 4D Gaussian Transformer Using Real-World Monocular Videos (NeurIPS 2025 · similarity 0.90)
- Flux4D: Flow-based Unsupervised 4D Reconstruction (NeurIPS 2025 · similarity 0.89)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.88)
- IGGT4D: Streaming 4D Instance-Grounded Geometry Transformer (NeurIPS 2026 · similarity 0.88)
- 4D-LRM: Large Space-Time Reconstruction Model From and To Any View at Any Time (NeurIPS 2025 · similarity 0.87)
- Motion4D: Learning 3D-Consistent Motion and Semantics for 4D Scene Understanding (NeurIPS 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.
