Research map / Video / Video generation
Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Sampling
Qi Sun, Can Wang, Jiaxiang Shang, Yingchun Liu, Jing Liao
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 |
|---|---|---|---|
| use self-guided stochastic sampling with layered motion instead of LHM kinematic animation | Ani3DHuman vs LHM · ActorsHQ | FID: 124.1 -> 105.3 | Improved |
| use Ani3DHuman instead of LHM | Ani3DHuman vs LHM · ActorsHQ | SSIM: 0.8382 -> 0.8312 | Hurt |
| use Ani3DHuman instead of LHM | Ani3DHuman vs LHM · ActorsHQ | FVD: 339.9 -> 295.2 | Improved |
| use Ani3DHuman instead of LHM | Ani3DHuman vs LHM · User study | Non-rigid physical plausibility (%): 14.7 -> 61.8 | Improved |
| replace deterministic ODE with stochastic sampling (ablation) | self-guided sampler · video re-rendering (qualitative) | visual quality | Improved |
| add self-guidance term (ablation) | stochastic sampler · video re-rendering (qualitative) | identity preservation | Improved |
| use personalized human diffusion prior instead of general prior (ablation) | self-guided sampler · video re-rendering (qualitative) | realism | Improved |
| use layered motion instead of single-layer motion field (ablation) | Ani3DHuman · 4D optimization (qualitative) | hand detail | Improved |
| use diagonal view-time sampling instead of bullet-time or independent-view (ablation) | Ani3DHuman · 4D optimization (qualitative, Ntraj=3) | floaters and spikes | 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.
- SV4D 2.0 — SV4D 2.0: Enhancing Spatio-Temporal Consistency in Multi-View Video Diffusion for High-Quality 4D Generation (ICCV 2025) · better on 8 of 8 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
- Animate3D: Animating Any 3D Model with Multi-view Video Diffusion (NeurIPS 2024 · similarity 0.89)
- Human-3Diffusion: Realistic Avatar Creation via Explicit 3D Consistent Diffusion Models (NeurIPS 2024 · similarity 0.88)
- PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025 · similarity 0.87)
- DiffHuman: Probabilistic Photorealistic 3D Reconstruction of Humans (CVPR 2024 · similarity 0.87)
- MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model (CVPR 2024 · similarity 0.87)
- Align Your Gaussians: Text-to-4D with Dynamic 3D Gaussians and Composed Diffusion Models (CVPR 2024 · 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.
