Research map / 3D vision / Human pose estimation
TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos
Yufu Wang, Ziyun Wang, Lingjie Liu, Kostas Daniilidis
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 |
|---|---|---|---|
| mask dynamic humans in both the input image and the dense bundle adjustment of DROID-SLAM (ablation) | DROID-SLAM · EMDB 2 camera trajectory | ATE average (m): 2.42 -> 0.32 | Improved |
| mask input images only (ablation) | DROID-SLAM · EMDB 2 camera trajectory | ATE average (m): 2.42 -> 1.42 | Improved |
| mask confidence in dense bundle adjustment only (ablation) | DROID-SLAM · EMDB 2 camera trajectory | ATE average (m): 2.42 -> 0.91 | Improved |
| use robust per-keyframe scale estimation with median instead of feeding ZoeDepth into DROID (ablation) | Masked DROID · EMDB 2 camera trajectory | ATE-S average (m): 3.09 -> 0.66 | Improved |
| compose camera trajectory from masked metric SLAM with camera-frame human motion instead of regressing trajectory | TRAM · EMDB 2 | RTE (%): 4.6 -> 1.4 | Improved |
| add token temporal transformer to HMR2.0(ft) (ablation) | VIMO · EMDB body motion | Accel (m/s^2): 20.5 -> 11.6 | Improved |
| add motion temporal transformer to HMR2.0(ft) (ablation) | VIMO · EMDB body motion | Accel (m/s^2): 20.5 -> 5.2 | Improved |
| add both temporal transformers (VIMO) to finetuned HMR2.0 (ablation) | HMR2.0(ft) · 3DPW body motion | Accel (m/s^2): 14.8 -> 4.9 | Improved |
| use ORB-SLAM2 as base instead of DROID-SLAM | SLAM camera estimation · EMDB 2 | robustness | Hurt |
| scale estimation versus ground truth scale | Masked DROID · EMDB 2 | ATE average (m): 0.32 -> 0.66 | Mixed |
Later papers that compared with this method
8 later accepted papers report TRAM as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.
- HumanBA: Human-Aware Bundle Adjustment via Global Human-Camera Decoupling (CVPR 2026 · better on 7 of 9 shared results)
- FactorizedHMR: A Hybrid Framework for Video Human Mesh Recovery (NeurIPS 2026 · better on 2 of 3 shared results)
- UniSH: Unifying Scene and Human Reconstruction in a Feed-Forward Pass (CVPR 2026 · better on 1 of 3 shared results)
- MetricHMSR: Metric Human Mesh and Scene Recovery from Monocular Images (CVPR 2026 · better on 3 of 3 shared results)
- Joint Optimization for 4D Human-Scene Reconstruction in the Wild (ICLR 2026 · better on 2 of 2 shared results)
- Contact-guided Real2Sim from Monocular Video with Planar Scene Primitives (ICLR 2026 · better on 1 of 1 shared result)
- Human3R: Everyone Everywhere All at Once (ICLR 2026 · better on 0 of 1 shared result)
- PromptHMR: Promptable Human Mesh Recovery (CVPR 2025 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DROID-SLAM — DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras (NeurIPS 2021) · better on 1 of 1 shared result
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
- Decoupling Human and Camera Motion From Videos in the Wild (CVPR 2023 · similarity 0.92)
- HumanMM: Global Human Motion Recovery from Multi-shot Videos (CVPR 2025 · similarity 0.88)
- WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion (CVPR 2024 · similarity 0.88)
- Trajectory Optimization for Physics-Based Reconstruction of 3D Human Pose From Monocular Video (CVPR 2022 · similarity 0.87)
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.85)
- Synergistic Global-space Camera and Human Reconstruction from Videos (CVPR 2024 · similarity 0.85)
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.
