Research map / 3D vision / 3D generation
TROPHIES: Temporal Reconstruction of Places, Humans, and Cameras from Multi-view Videos
Jinpeng Liu, Yukang Xu, Yutong Li, Xingyu Liu
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 human-aware attention to scene branch (ablation) | DUSt3R · EgoHumans | TE: 3.20 -> 3.05 | Improved |
| add human-aware attention to scene branch (ablation) | MonST3R · EgoHumans | TE: 3.17 -> 3.06 | Improved |
| add human-aware attention to scene branch (ablation) | CUT3R · EgoHumans | TE: 1.90 -> 1.83 | Improved |
| add human-aware attention to scene branch (ablation) | DUSt3R · EgoHumans | s-CCA@100: 0.49 -> 0.52 | Improved |
| add human-aware attention to scene branch (ablation) | MonST3R · EgoHumans | AE: 108.82 -> 101.71 | Improved |
| add human-aware attention to scene branch (ablation) | CUT3R · EgoHumans | s-CCA@100: 0.48 -> 0.52 | Improved |
| add gravity-aware term to human branch (ablation) | Trophies human branch · EgoHumans (All Views) | MPJPE: 79.1 -> 77.7 | Improved |
| add gravity-aware term to human branch (ablation) | Trophies human branch · EgoHumans (Anchor View) | Accel: 17.3 -> 13.6 | Improved |
| add symmetric and anchor-referenced cross-view attention (no gravity) over fine-tuned VIMO | VIMO · EgoHumans (All Views) | MPJPE: 81.2 -> 79.1 | Improved |
| replace per-frame HSfM with global alignment and optimization | HSfM · EgoHumans | W-MPJPE: 227.82 -> 97.54 | Improved |
| apply Trophies to DUSt3R | DUSt3R · EgoExo4D | TE: 4.84 -> 2.37 | Improved |
| apply Trophies to DUSt3R | DUSt3R · EgoHumans | PA-MPJPE: 21.93 -> 22.74 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MonST3R — MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion (ICLR 2025) · better on 2 of 2 shared results
- DUSt3R — DUSt3R: Geometric 3D Vision Made Easy (CVPR 2024) · better on 2 of 2 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
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.92)
- Synergistic Global-space Camera and Human Reconstruction from Videos (CVPR 2024 · similarity 0.90)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.89)
- HumanMM: Global Human Motion Recovery from Multi-shot Videos (CVPR 2025 · similarity 0.88)
- Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses (CVPR 2024 · similarity 0.88)
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video (NeurIPS 2026 · 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.
