Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2602.19350 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914343504838656 |
|---|---|
| author | Guo, Zhilin Yang, Jing Fogarty, Kyle Wan, Jingyi Zhang, Boqiao Wu, Tianhao Xia, Weihao Zhou, Chenliang Khattar, Sakar Zhong, Fangcheng Vasconcelos, Cristina Nader Oztireli, Cengiz |
| author_facet | Guo, Zhilin Yang, Jing Fogarty, Kyle Wan, Jingyi Zhang, Boqiao Wu, Tianhao Xia, Weihao Zhou, Chenliang Khattar, Sakar Zhong, Fangcheng Vasconcelos, Cristina Nader Oztireli, Cengiz |
| contents | Digitizing humans and synthesizing photorealistic avatars with explicit 3D pose and camera controls are central to VR, telepresence, and entertainment. Existing skinning-based workflows require laborious manual rigging or template-based fittings, while neural volumetric methods rely on canonical templates and re-optimization for each unseen pose. We present PoseCraft, a diffusion framework built around tokenized 3D interface: instead of relying only on rasterized geometry as 2D control images, we encode sparse 3D landmarks and camera extrinsics as discrete conditioning tokens and inject them into diffusion via cross-attention. Our approach preserves 3D semantics by avoiding 2D re-projection ambiguity under large pose and viewpoint changes, and produces photorealistic imagery that faithfully captures identity and appearance. To train and evaluate at scale, we also implement GenHumanRF, a data generation workflow that renders diverse supervision from volumetric reconstructions. Our experiments show that PoseCraft achieves significant perceptual quality improvement over diffusion-centric methods, and attains better or comparable metrics to latest volumetric rendering SOTA while better preserving fabric and hair details. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_19350 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PoseCraft: Tokenized 3D Body Landmark and Camera Conditioning for Photorealistic Human Image Synthesis Guo, Zhilin Yang, Jing Fogarty, Kyle Wan, Jingyi Zhang, Boqiao Wu, Tianhao Xia, Weihao Zhou, Chenliang Khattar, Sakar Zhong, Fangcheng Vasconcelos, Cristina Nader Oztireli, Cengiz Computer Vision and Pattern Recognition Digitizing humans and synthesizing photorealistic avatars with explicit 3D pose and camera controls are central to VR, telepresence, and entertainment. Existing skinning-based workflows require laborious manual rigging or template-based fittings, while neural volumetric methods rely on canonical templates and re-optimization for each unseen pose. We present PoseCraft, a diffusion framework built around tokenized 3D interface: instead of relying only on rasterized geometry as 2D control images, we encode sparse 3D landmarks and camera extrinsics as discrete conditioning tokens and inject them into diffusion via cross-attention. Our approach preserves 3D semantics by avoiding 2D re-projection ambiguity under large pose and viewpoint changes, and produces photorealistic imagery that faithfully captures identity and appearance. To train and evaluate at scale, we also implement GenHumanRF, a data generation workflow that renders diverse supervision from volumetric reconstructions. Our experiments show that PoseCraft achieves significant perceptual quality improvement over diffusion-centric methods, and attains better or comparable metrics to latest volumetric rendering SOTA while better preserving fabric and hair details. |
| title | PoseCraft: Tokenized 3D Body Landmark and Camera Conditioning for Photorealistic Human Image Synthesis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.19350 |