Research map / Video / Video generation
ParCo: Part-Coordinating Text-to-Motion Synthesis
Qiran Zou, Shangyuan Yuan, Shian Du, Yu Wang, Chang Liu, Yi Xu, Jie Chen, Xiangyang Ji
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 part coordination module to 6-part generators (ablation) | ParCo · HumanML3D test | FID: 3.652 -> 0.109 | Improved |
| add part coordination module to upper/lower-body generators (ablation) | Up&LowBody discretization · HumanML3D test | FID: 0.497 -> 0.172 | Improved |
| add part coordination module to 6-part generators (ablation) | ParCo · HumanML3D test | R-Precision Top-1: 0.375 -> 0.515 | Improved |
| use 6-part partition instead of upper/lower body, with coordination (ablation) | ParCo · HumanML3D test | FID: 0.172 -> 0.109 | Improved |
| use 6 parts without coordination instead of 2 parts without coordination (ablation) | part generators without coordination · HumanML3D test | FID: 0.497 -> 3.652 | Hurt |
| evaluate ReMoDiffuse with random motion length instead of ground-truth length (ablation) | ReMoDiffuse · HumanML3D test | FID: 0.103 -> 0.281 | Hurt |
| evaluate ReMoDiffuse with random motion length instead of ground-truth length (ablation) | ReMoDiffuse · KIT-ML test | FID: 0.155 -> 0.589 | Hurt |
| use ParCo instead of T2M-GPT across text-length quartiles | text-to-motion · HumanML3D test subsets by text length | metrics across quartiles | Improved |
| use six small part generators instead of a single large generator | T2M-GPT baseline · HumanML3D test | parameters (M) / inference time (s): 237.6 -> 168.4 M; 0.544 -> 0.036 s | Improved |
| use ParCo's 6 part VQ-VAEs instead of a single whole-body VQ-VAE (ablation) | motion VQ-VAE · HumanML3D reconstruction | FID: 0.070 -> 0.021 | Improved |
| ParCo vs AttT2M on diversity of generation for one text | ParCo · HumanML3D test | MModality: 2.452 -> 1.382 | Hurt |
Later papers that compared with this method
3 later accepted papers report ParCo as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- ParTY: Part-Guidance for Expressive Text-to-Motion Synthesis (CVPR 2026 · better on 7 of 7 shared results)
- Next-Scale Autoregressive Models for Text-to-Motion Generation (CVPR 2026 · better on 4 of 4 shared results)
- K-Mask: Kinematic-Aware Masked Modeling for Controllable Text-to-Motion Synthesis (ECCV 2026 · 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 2 of 2 on most shared results.
- AttT2M — AttT2M: Text-Driven Human Motion Generation with Multi-Perspective Attention Mechanism (ICCV 2023) · better on 2 of 3 shared results
- Fg-T2M — Fg-T2M: Fine-Grained Text-Driven Human Motion Generation via Diffusion Model (ICCV 2023) · 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
- ParTY: Part-Guidance for Expressive Text-to-Motion Synthesis (CVPR 2026 · similarity 0.92)
- FrankenMotion: Part-level Human Motion Generation and Composition (CVPR 2026 · similarity 0.88)
- KinMo: Kinematic-aware Human Motion Understanding and Generation (ICCV 2025 · similarity 0.88)
- How to Move Your Dragon: Text-to-Motion Synthesis for Large-Vocabulary Objects (ICML 2025 · similarity 0.87)
- HumanTOMATO: Text-aligned Whole-body Motion Generation (ICML 2024 · similarity 0.87)
- OMG: Towards Open-vocabulary Motion Generation via Mixture of Controllers (CVPR 2024 · similarity 0.86)
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.
