Research map / 3D vision / 3D generation
UniEgoMotion: A Unified Model for Egocentric Motion Reconstruction, Forecasting, and Generation
Chaitanya Patel, Hiroki Nakamura, Yuta Kyuragi, Kazuki Kozuka, Juan Carlos Niebles, Ehsan Adeli
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 |
|---|---|---|---|
| use head-centric instead of pelvis-centric motion representation (ablation) | UniEgoMotion · EE4D-Motion reconstruction | MPJPE: 0.166 -> 0.100 | Improved |
| use head-centric instead of pelvis-centric motion representation (ablation) | UniEgoMotion · EE4D-Motion reconstruction | Head Trans. Err.: 0.138 -> 0.058 | Improved |
| use head-centric instead of global motion representation (ablation) | UniEgoMotion · EE4D-Motion forecasting | MPJPE 2-4s: 0.293 -> 0.206 | Improved |
| use head-centric instead of global motion representation (ablation) | UniEgoMotion · EE4D-Motion reconstruction | Head Trans. Err.: 0.051 -> 0.058 | Hurt |
| use transformer decoder instead of transformer encoder (ablation) | UniEgoMotion · EE4D-Motion reconstruction | Foot Slide: 4.48 -> 3.62 | Improved |
| use transformer decoder instead of 1D U-Net (ablation) | UniEgoMotion · EE4D-Motion reconstruction | MPJPE: 0.145 -> 0.100 | Improved |
| use transformer decoder instead of 1D U-Net (ablation) | UniEgoMotion · EE4D-Motion generation | MPJPE 0-2s: 0.259 -> 0.226 | Improved |
| use DINOv2 instead of CLIP image encoder (ablation) | UniEgoMotion · EE4D-Motion reconstruction | Foot Slide: 4.03 -> 3.62 | Improved |
| use DINOv2 instead of EgoVideo image encoder (ablation) | UniEgoMotion · EE4D-Motion reconstruction | MPJPE: 0.132 -> 0.100 | Improved |
| use DINOv2 instead of EgoVideo image encoder (ablation) | UniEgoMotion · EE4D-Motion forecasting | FID: 0.060 -> 0.047 | Improved |
| use DINOv2 instead of CLIP image encoder (ablation) | UniEgoMotion · EE4D-Motion generation | Foot Slide: 3.57 -> 2.89 | Improved |
| use transformer decoder instead of transformer encoder (ablation) | UniEgoMotion · EE4D-Motion reconstruction | FID: 0.017 -> 0.027 | Hurt |
Later papers that compared with this method
1 later accepted paper reports UniEgoMotion as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Ego-Human Motion Prediction with 3D-Aware LLM (ECCV 2026 · better on 4 of 4 shared results)
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.
- AvatarPoser — AvatarPoser: Articulated Full-Body Pose Tracking from Sparse Motion Sensing (ECCV 2022) · better on 3 of 3 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
- EgoGen: An Egocentric Synthetic Data Generator (CVPR 2024 · similarity 0.92)
- Gaze Beyond the Frame: Forecasting Egocentric 3D Visual Span (NeurIPS 2025 · similarity 0.91)
- EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity (NeurIPS 2024 · similarity 0.88)
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.88)
- PlayerOne: Egocentric World Simulator (NeurIPS 2025 · similarity 0.88)
- Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation (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.
