![[2602.19089] Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Sampling](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.19089] Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Sampling
Summary
Ani3DHuman presents a novel framework for photorealistic 3D human animation, combining kinematics-based methods with video diffusion priors to enhance motion realism and quality.
Why It Matters
This research addresses significant challenges in 3D human animation, such as achieving photorealism and maintaining identity fidelity. By introducing a self-guided stochastic sampling approach, it advances the field of computer vision and graphics, potentially impacting industries like gaming, film, and virtual reality.
Key Takeaways
- Ani3DHuman integrates kinematics with video diffusion for enhanced animation quality.
- The framework overcomes limitations of existing methods, achieving photorealistic results.
- A novel self-guided stochastic sampling method addresses out-of-distribution challenges.
- Extensive experiments validate the effectiveness of the proposed approach.
- Code availability encourages further research and application in the field.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.19089 (cs) [Submitted on 22 Feb 2026] Title:Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Sampling Authors:Qi Sun, Can Wang, Jiaxiang Shang, Yingchun Liu, Jing Liao View a PDF of the paper titled Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Sampling, by Qi Sun and 4 other authors View PDF HTML (experimental) Abstract:Current 3D human animation methods struggle to achieve photorealism: kinematics-based approaches lack non-rigid dynamics (e.g., clothing dynamics), while methods that leverage video diffusion priors can synthesize non-rigid motion but suffer from quality artifacts and identity loss. To overcome these limitations, we present Ani3DHuman, a framework that marries kinematics-based animation with video diffusion priors. We first introduce a layered motion representation that disentangles rigid motion from residual non-rigid motion. Rigid motion is generated by a kinematic method, which then produces a coarse rendering to guide the video diffusion model in generating video sequences that restore the residual non-rigid motion. However, this restoration task, based on diffusion sampling, is highly challenging, as the initial renderings are out-of-distribution, causing standard deterministic ODE samplers to fail. Therefore, we propose a novel self-guided stochastic sampling method, which effectively addresses the out-of-distribution problem by ...