Research map / Video / Video generation
Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordance
Li Hu, Guangyuan Wang, Zhen Shen, Xin Gao, Dechao Meng, Lian Zhuo, Peng Zhang, Bang Zhang, Liefeng Bo
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 spatial blending of object guider features (ablation) | Animate Anyone 2 · Own dataset (ablation) | FVD: 191.23 -> 186.17 | Improved |
| Add spatial blending of object guider features (ablation) | Animate Anyone 2 · Own dataset (ablation) | SSIM: 0.789 -> 0.794 | Improved |
| Add depth-wise pose modulation (ablation) | Animate Anyone 2 · Own dataset (ablation) | FVD: 211.15 -> 186.17 | Improved |
| Add depth-wise pose modulation (ablation) | Animate Anyone 2 · Own dataset (ablation) | SSIM: 0.769 -> 0.794 | Improved |
| Use shape-agnostic mask instead of precise character mask or bounding box (ablation) | Animate Anyone 2 · qualitative ablation | visual quality | Improved |
| Use the full environment-aware model instead of compositing animation onto inpainted background | Animate Anyone 2 · Own dataset (character-environment affordance) | FVD: 195.45 -> 186.17 | Improved |
| Add extra video pretraining before fine-tuning on TikTok | Animate Anyone 2 · TikTok benchmark | FVD: 158.97 -> 144.65 | Improved |
| Add extra video pretraining before fine-tuning on TikTok | Animate Anyone 2 · TikTok benchmark | SSIM: 0.778 -> 0.812 | Improved |
| Use Animate Anyone 2 instead of Animate Anyone (TikTok only) | Animate Anyone · TikTok benchmark | FVD: 171.90 -> 158.97 | Improved |
| Use Animate Anyone 2 instead of Champ on own dataset | Champ · Own dataset (100 videos) | FVD: 205.79 -> 172.54 | Improved |
| Use Animate Anyone 2 without extra pretraining instead of Champ* with it | Champ* · TikTok benchmark | SSIM: 0.802 -> 0.778 | Hurt |
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.
- Animate Anyone — Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animation (CVPR 2024) · better on 4 of 4 shared results
- MimicMotion — MimicMotion: High-Quality Human Motion Video Generation with Confidence-aware Pose Guidance (ICML 2025) · better on 2 of 2 shared results
- MagicAnimate — MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model (CVPR 2024) · better on 2 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
- Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animation (CVPR 2024 · similarity 0.92)
- AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion Models (CVPR 2025 · similarity 0.87)
- Animate-X: Universal Character Image Animation with Enhanced Motion Representation (ICLR 2025 · similarity 0.87)
- Free-viewpoint Human Animation with Pose-correlated Reference Selection (CVPR 2025 · similarity 0.87)
- AniDoc: Animation Creation Made Easier (CVPR 2025 · similarity 0.85)
- Towards Multiple Character Image Animation Through Enhancing Implicit Decoupling (ICLR 2025 · similarity 0.85)
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.
