Research map / Video / Video generation
MotionCLIP: Exposing Human Motion Generation to CLIP Space
Guy Tevet, Brian Gordon, Amir Hertz, Amit H. Bermano, Daniel Cohen-Or
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 image loss (ablation) | MotionCLIP · BABEL-60 action recognition | top-1 accuracy: 35.05% -> 40.9% | Improved |
| add image loss (ablation) | MotionCLIP · BABEL-60 action recognition | top-5 accuracy: 50.26% -> 57.71% | Improved |
| add text loss (ablation) | MotionCLIP · BABEL-60 action recognition | top-1 accuracy: 4.54% -> 40.9% | Improved |
| add text loss (ablation) | MotionCLIP · BABEL-60 action recognition | top-5 accuracy: 18.37% -> 57.71% | Improved |
| use MotionCLIP instead of JL2P for text-to-motion | text-to-motion · Action generation user study (in-domain set) | user preference: 23.3% -> 76.7% | Improved |
| use MotionCLIP instead of JL2P for unseen actions | text-to-motion · Action generation user study (out-of-domain set) | user preference: 24.7% -> 75.3% | Improved |
| use text-only MotionCLIP instead of Aberman et al. style transfer | style generation · Style generation user study | user preference: 62.1% -> 37.9% | Hurt |
| use MotionCLIP instead of 2s-AGCN for classification | action recognition · BABEL-60 action recognition | top-1 accuracy: 41.14% -> 40.9% | No clear effect |
| use MotionCLIP instead of 2s-AGCN for classification | action recognition · BABEL-60 action recognition | top-5 accuracy: 73.18% -> 57.71% | Hurt |
Later papers that compared with this method
3 later accepted papers report MotionCLIP as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- "Plan, Posture and Go: Towards Open-vocabulary Text-to-Motion Generation" (ECCV 2024 · better on 3 of 3 shared results)
- FreeMotion: MoCap-Free Human Motion Synthesis with Multimodal Large Language Models (ECCV 2024 · better on 1 of 1 shared result)
- UDE: A Unified Driving Engine for Human Motion Generation (CVPR 2023 · better on 0 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
- Enhanced Motion-Text Alignment for Image-to-Video Transfer Learning (CVPR 2024 · similarity 0.89)
- LaMP: Language-Motion Pretraining for Motion Generation, Retrieval, and Captioning (ICLR 2025 · similarity 0.88)
- MotionGPT3: Human Motion as a Second Modality (ICLR 2026 · similarity 0.88)
- SimMotionEdit: Text-Based Human Motion Editing with Motion Similarity Prediction (CVPR 2025 · similarity 0.88)
- Tri-Modal Motion Retrieval by Learning a Joint Embedding Space (CVPR 2024 · similarity 0.87)
- SnapMoGen: Human Motion Generation from Expressive Texts (NeurIPS 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.
