Research map / Video / Video generation
Multi-identity Human Image Animation with Structural Video Diffusion
Zhenzhi Wang, Yixuan Li, Yanhong Zeng, Yuwei Guo, Dahua Lin, Tianfan Xue, Bo Dai
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 ID-embedding (ablation) | CamAnimate baseline · Multi-ID test set | FID: 54.09 -> 33.75 | Improved |
| add ID-embedding (ablation) | CamAnimate baseline · Multi-ID test set | PSNR: 19.552 -> 20.374 | Improved |
| add ID-embedding (ablation) | CamAnimate baseline · Multi-ID test set | FVD: 982.1 -> 873.5 | Improved |
| add depth/multi-modality prediction (ablation) | CamAnimate baseline · Multi-ID test set | FID: 54.09 -> 47.67 | Improved |
| add depth/multi-modality prediction (ablation) | CamAnimate baseline · Multi-ID test set | SSIM: 0.649 -> 0.668 | Improved |
| add depth prediction on top of ID-embedding (ablation) | ID-embedding model · Multi-ID test set | FID: 33.75 -> 30.57 | Improved |
| add depth prediction on top of ID-embedding (ablation) | ID-embedding model · Multi-ID test set | FVD: 873.5 -> 878.2 | Hurt |
| add normal prediction on top of ID-embedding (ablation) | ID-embedding model · Multi-ID test set | FID: 33.75 -> 60.58 | Hurt |
| add normal prediction on top of ID-embedding (ablation) | ID-embedding model · Multi-ID test set | PSNR: 20.374 -> 19.037 | Hurt |
| add normal prediction alongside depth (ablation) | ID-embedding + depth model · Multi-ID test set | FID: 30.57 -> 56.78 | Hurt |
| full method vs CamAnimate | Structural Video Diffusion · Multi-ID test set | FID: 54.09 -> 30.57 | Improved |
| full method vs CamAnimate | Structural Video Diffusion · Multi-ID test set | FVD: 982.1 -> 878.2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MimicMotion — MimicMotion: High-Quality Human Motion Video Generation with Confidence-aware Pose Guidance (ICML 2025) · better on 5 of 5 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 0.88)
- Human Motion Diffusion as a Generative Prior (ICLR 2024 · similarity 0.88)
- ShowMaker: Creating High-Fidelity 2D Human Video via Fine-Grained Diffusion Modeling (NeurIPS 2024 · similarity 0.88)
- StableAnimator: High-Quality Identity-Preserving Human Image Animation (CVPR 2025 · similarity 0.88)
- Move-in-2D: 2D-Conditioned Human Motion Generation (CVPR 2025 · similarity 0.88)
- VLOGGER: Multimodal Diffusion for Embodied Avatar Synthesis (CVPR 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.
