Research map / 3D vision / Human pose estimation
Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidance
Shenhao Zhu, Junming Leo Chen, Zuozhuo Dai, Zilong Dong, Yinghui Xu, Xun Cao, Yao Yao, Hao Zhu, Siyu Zhu
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 SMPL-rendered depth, normal and semantic maps to skeleton guidance (ablation) | Champ · TikTok | FVD: 192.34 -> 170.20 | Improved |
| add depth and normal geometry maps to the motion condition (ablation) | Champ · TikTok | FVD: 189.07 -> 170.20 | Improved |
| add skeleton guidance on top of SMPL conditions (ablation) | Champ · TikTok | PSNR: 29.05 -> 29.84 | Improved |
| add guidance self-attention in the motion fusion module (ablation) | Champ · TikTok | FVD: 174.46 -> 170.20 | Improved |
| fine-tune on the TikTok training set | Champ · TikTok | FVD: 170.20 -> 160.82 | Improved |
| use SMPL-based shape alignment instead of baselines on an unseen dataset | Champ · Proposed unseen in-the-wild dataset | FVD: 193.6 -> 173.5 (vs MagicAnimate) | Improved |
| use non-fine-tuned Champ rather than MagicAnimate on TikTok FID-VID | Champ · TikTok | FID-VID: 21.75 -> 26.14 | Hurt |
Later papers that compared with this method
8 later accepted papers report Champ as a baseline in their results tables; 8 of them report a better number on most of the results they share with it.
- Feed-forward Human Performance Capture via Progressive Canonical Space Updates (ICLR 2026 · better on 6 of 6 shared results)
- DreamDance: Animating Human Images by Enriching 3D Geometry Cues from 2D Poses (ICCV 2025 · better on 6 of 6 shared results)
- GAS: Generative Avatar Synthesis from a Single Image (ICCV 2025 · better on 6 of 6 shared results)
- MIMO: Controllable Character Video Synthesis with Spatial Decomposed Modeling (CVPR 2025 · better on 4 of 4 shared results)
- GUAVA: Generalizable Upper Body 3D Gaussian Avatar (ICCV 2025 · better on 4 of 4 shared results)
- AniGS: Animatable Gaussian Avatar from a Single Image with Inconsistent Gaussian Reconstruction (CVPR 2025 · better on 3 of 3 shared results)
- PERSONA: Personalized Whole-Body 3D Avatar with Pose-Driven Deformations from a Single Image (ICCV 2025 · better on 2 of 3 shared results)
- MagicMan: Generative Novel View Synthesis of Humans with 3D-Aware Diffusion and Iterative Refinement (AAAI 2025 · 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
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.89)
- DreamDance: Animating Human Images by Enriching 3D Geometry Cues from 2D Poses (ICCV 2025 · similarity 0.89)
- Human Geometry Distribution for 3D Animation Generation (CVPR 2026 · similarity 0.89)
- Action2video: Generating Videos of Human 3D Actions (IJCV 2022 · similarity 0.89)
- DisPose: Disentangling Pose Guidance for Controllable Human Image Animation (ICLR 2025 · similarity 0.88)
- Free-viewpoint Human Animation with Pose-correlated Reference Selection (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.
