Research map / Video / Video generation
GENMO: A GENeralist Model for Human MOtion
Jiefeng Li, Jinkun Cao, Haotian Zhang, Davis Rempe, Jan Kautz, Umar Iqbal, Ye Yuan
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 2D-only (Motion-X) training via estimation-guided objective (ablation) | GENMO · 3DPW camera-space | MPJPE: 35.2 -> 34.6 | Improved |
| add 2D-only (Motion-X) training (ablation) | GENMO · Motion-X text-to-motion | FID: 0.515 -> 0.207 | Improved |
| add 2D-only (Motion-X) training (ablation) | GENMO · HumanML3D text-to-motion | R@3: 0.556 -> 0.632 | Improved |
| add 2D-only training (ablation) | GENMO · HumanML3D text-to-motion | MM Dist: 3.128 -> 3.466 | Hurt |
| add estimation mode to diffusion training (ablation) | GENMO · EMDB (24) global motion | W-MPJPE100: 307.7 -> 281.7 | Improved |
| add estimation mode to diffusion training (ablation) | GENMO · RICH (24) global motion | W-MPJPE100: 143.9 -> 130.6 | Improved |
| add generation mode (generative prior) to regression-only training (ablation) | GENMO · EMDB (24) global motion | W-MPJPE100: 300.1 -> 281.7 | Improved |
| add generation mode to regression-only training (ablation) | GENMO · RICH (24) global motion | WA-MPJPE100: 87.0 -> 81.3 | Improved |
| add generation mode to regression-only training (ablation) | GENMO · 3DPW-XOCC occlusion | MPJPE: 89.0 -> 76.2 | Improved |
| add estimation mode to diffusion training for in-betweening (ablation) | GENMO · HumanML3D in-betweening (2 keyframes) | PA-MPJPE: 64.9 -> 53.5 | Improved |
| add 2D training for in-betweening (ablation) | GENMO · Motion-X in-betweening (2 keyframes) | PA-MPJPE: 68.3 -> 58.8 | Improved |
| increase DDIM inference steps from 1 to 5 (ablation) | GENMO · EMDB estimation | W-MPJPE100: 280.0 -> 274.9 | Improved |
Later papers that compared with this method
3 later accepted papers report GENMO as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- DuoMo: Dual Motion Diffusion for World-Space Human Reconstruction (CVPR 2026 · better on 6 of 6 shared results)
- Odoriko: A Shape-Aware Multimodal Diffusion Framework for Human Motion (ECCV 2026 · better on 4 of 6 shared results)
- Video Generation Models are General-Purpose Vision Learners (ECCV 2026 · 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
- The Quest for Generalizable Motion Generation: Data, Model, and Evaluation (ICLR 2026 · similarity 0.91)
- KinMo: Kinematic-aware Human Motion Understanding and Generation (ICCV 2025 · similarity 0.90)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.88)
- Human Motion Diffusion as a Generative Prior (ICLR 2024 · similarity 0.88)
- ReMoGPT: Part-Level Retrieval-Augmented Motion-Language Models (AAAI 2025 · similarity 0.88)
- Scaling Large Motion Models with Million-Level Human Motions (ICML 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.
