Research map / Video / Video generation
Human-Centric Video Generation via Collaborative Multi-Modal Conditioning
Liyang Chen, Tianxiang Ma, Jiawei Liu, Bingchuan Li, Zhuowei Chen, Lijie Liu, Xu He, Gen Li, Qian He, Zhiyong Wu
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 |
|---|---|---|---|
| confine finetuning to self-attention layers instead of full finetuning (ablation) | HuMo-17B · MoCha | AES: 0.529 -> 0.589 | Improved |
| confine finetuning to self-attention layers instead of full finetuning (ablation) | HuMo-17B · MoCha | TVA: 6.157 -> 6.508 | Improved |
| use progressive two-stage training instead of single-stage joint training (ablation) | HuMo-17B · MoCha | ID-Cur: 0.724 -> 0.747 | Improved |
| use progressive training (ablation) | HuMo-17B · MoCha | AES: 0.541 -> 0.589 | No clear effect |
| add face-location prediction (focus-by-predicting) (ablation) | HuMo-17B · MoCha | Sync-C: 5.946 -> 6.252 | Improved |
| add face-location prediction (ablation) | HuMo-17B · MoCha | ID-Cur: 0.730 -> 0.747 | Improved |
| add face-location prediction (ablation) | HuMo-17B · MoCha | TVA: 6.507 -> 6.508 | No clear effect |
| scale model from 1.7B to 17B | HuMo · In-house subject preservation | TVA: 3.222 -> 3.939 | Improved |
| HuMo-17B vs OmniHuman-1 on lip sync | HuMo-17B · MoCha | Sync-C: 6.526 vs 6.252 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- OmniHuman-1 — OmniHuman-1: Rethinking the Scaling-Up of One-Stage Conditioned Human Animation Models (ICCV 2025) · better on 0 of 2 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
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.87)
- OpenHumanVid: A Large-Scale High-Quality Dataset for Enhancing Human-Centric Video Generation (CVPR 2025 · similarity 0.87)
- VLOGGER: Multimodal Diffusion for Embodied Avatar Synthesis (CVPR 2025 · similarity 0.87)
- MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis (CVPR 2025 · similarity 0.86)
- Let Them Talk: Audio-Driven Multi-Person Conversational Video Generation (NeurIPS 2025 · similarity 0.86)
- C3Net: Compound Conditioned ControlNet for Multimodal Content Generation (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.
