AdaHuman: Animatable Detailed 3D Human Generation with Compositional Multiview Diffusion
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912404341784576 |
|---|---|
| author | Huang, Yangyi Yuan, Ye Li, Xueting Kautz, Jan Iqbal, Umar |
| author_facet | Huang, Yangyi Yuan, Ye Li, Xueting Kautz, Jan Iqbal, Umar |
| contents | Existing methods for image-to-3D avatar generation struggle to produce highly detailed, animation-ready avatars suitable for real-world applications. We introduce AdaHuman, a novel framework that generates high-fidelity animatable 3D avatars from a single in-the-wild image. AdaHuman incorporates two key innovations: (1) A pose-conditioned 3D joint diffusion model that synthesizes consistent multi-view images in arbitrary poses alongside corresponding 3D Gaussian Splats (3DGS) reconstruction at each diffusion step; (2) A compositional 3DGS refinement module that enhances the details of local body parts through image-to-image refinement and seamlessly integrates them using a novel crop-aware camera ray map, producing a cohesive detailed 3D avatar. These components allow AdaHuman to generate highly realistic standardized A-pose avatars with minimal self-occlusion, enabling rigging and animation with any input motion. Extensive evaluation on public benchmarks and in-the-wild images demonstrates that AdaHuman significantly outperforms state-of-the-art methods in both avatar reconstruction and reposing. Code and models will be publicly available for research purposes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_24877 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AdaHuman: Animatable Detailed 3D Human Generation with Compositional Multiview Diffusion Huang, Yangyi Yuan, Ye Li, Xueting Kautz, Jan Iqbal, Umar Computer Vision and Pattern Recognition Existing methods for image-to-3D avatar generation struggle to produce highly detailed, animation-ready avatars suitable for real-world applications. We introduce AdaHuman, a novel framework that generates high-fidelity animatable 3D avatars from a single in-the-wild image. AdaHuman incorporates two key innovations: (1) A pose-conditioned 3D joint diffusion model that synthesizes consistent multi-view images in arbitrary poses alongside corresponding 3D Gaussian Splats (3DGS) reconstruction at each diffusion step; (2) A compositional 3DGS refinement module that enhances the details of local body parts through image-to-image refinement and seamlessly integrates them using a novel crop-aware camera ray map, producing a cohesive detailed 3D avatar. These components allow AdaHuman to generate highly realistic standardized A-pose avatars with minimal self-occlusion, enabling rigging and animation with any input motion. Extensive evaluation on public benchmarks and in-the-wild images demonstrates that AdaHuman significantly outperforms state-of-the-art methods in both avatar reconstruction and reposing. Code and models will be publicly available for research purposes. |
| title | AdaHuman: Animatable Detailed 3D Human Generation with Compositional Multiview Diffusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.24877 |