Research map / Video / Video generation
Odoriko: A Shape-Aware Multimodal Diffusion Framework for Human Motion
Dongseok Shim, Julian Tanke, Kengo Uchida, christian simon, Koichi Saito, Takashi Shibuya, Shusuke Takahashi, Yuki Mitsufuji
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 ground-truth shape conditioning (gender and beta) instead of canonical body (ablation) | Odoriko · HumanML3D | FID: 0.592 -> 0.103 | Improved |
| add ground-truth shape conditioning instead of canonical body (ablation) | Odoriko · HumanML3D | R-Precision Top-3: 0.768 -> 0.805 | Improved |
| use predicted gender and beta rather than neutral gender (ablation) | Odoriko · EMDB (24) | MPJPE: 76.3 -> 70.2 | Improved |
| use predicted beta rather than zero beta (ablation) | Odoriko · EMDB (24) | MPJPE: 72.7 -> 70.2 | Improved |
| use ground-truth shape instead of estimated shape (ablation) | Odoriko · EMDB (24) | MPJPE: 70.2 -> 70.0 | No clear effect |
| add camera-centric local root orientation (ablation) | Odoriko · EMDB (24) | MPJPE: 285.9 -> 70.2 | Improved |
| add 2D keypoint input (ablation) | Odoriko · 3DPW (14) | MPJPE: 100.2 -> 69.9 | Improved |
| add 2D keypoint input (ablation) | Odoriko · EMDB (24) | MPJPE: 71.1 -> 70.2 | No clear effect |
| increase UniPC denoising steps from 1 to 25 (ablation) | Odoriko · HumanML3D | FID: 1.804 -> 0.103 | Improved |
| increase UniPC steps from 25 to 50 (ablation) | Odoriko · HumanML3D | FID: 0.103 -> 0.113 | No clear effect |
| use UniPC sampler instead of DDIM at 25 steps (ablation) | Odoriko · HumanML3D | FID: 0.107 -> 0.103 | No clear effect |
| add CLIP global feature (ablation) | Odoriko · HumanML3D | R-Precision Top-3: 0.799 -> 0.805 | 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.
- GENMO — GENMO: A GENeralist Model for Human MOtion (ICCV 2025) · better on 4 of 6 shared results
- MotionCraft — MotionCraft: Physics-Based Zero-Shot Video Generation (NeurIPS 2024) · better on 5 of 5 shared results
- LODGE++ — Lodge++: High-Quality and Long Dance Generation With Robust Choreography Patterns (TPAMI 2025) · 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
- Shape My Moves: Text-Driven Shape-Aware Synthesis of Human Motions (CVPR 2025 · similarity 0.89)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.88)
- MotionMix: Weakly-Supervised Diffusion for Controllable Motion Generation (AAAI 2024 · similarity 0.88)
- 🎧MOSPA: Human Motion Generation Driven by Spatial Audio (NeurIPS 2025 · similarity 0.88)
- The Language of Motion: Unifying Verbal and Non-verbal Language of 3D Human Motion (CVPR 2025 · similarity 0.87)
- HuTuMotion: Human-Tuned Navigation of Latent Motion Diffusion Models with Minimal Feedback (AAAI 2024 · 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.
