Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2501.12202 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911684584538112 |
|---|---|
| author | Zhao, Zibo Lai, Zeqiang Lin, Qingxiang Zhao, Yunfei Liu, Haolin Yang, Shuhui Feng, Yifei Yang, Mingxin Zhang, Sheng Yang, Xianghui Shi, Huiwen Liu, Sicong Wu, Junta Lian, Yihang Yang, Fan Tang, Ruining He, Zebin Wang, Xinzhou Liu, Jian Zuo, Xuhui Chen, Zhuo Lei, Biwen Weng, Haohan Xu, Jing Zhu, Yiling Liu, Xinhai Xu, Lixin Hu, Changrong Yang, Shaoxiong Zhang, Song Liu, Yang Huang, Tianyu Wang, Lifu Zhang, Jihong Chen, Meng Dong, Liang Jia, Yiwen Cai, Yulin Yu, Jiaao Tang, Yixuan Zhang, Hao Ye, Zheng He, Peng Wu, Runzhou Zhang, Chao Tan, Yonghao Xiao, Jie Tao, Yangyu Zhu, Jianchen Xue, Jinbao Liu, Kai Zhao, Chongqing Wu, Xinming Hu, Zhichao Qin, Lei Peng, Jianbing Li, Zhan Chen, Minghui Zhang, Xipeng Niu, Lin Wang, Paige Wang, Yingkai Kuang, Haozhao Fan, Zhongyi Zheng, Xu Zhuang, Weihao He, YingPing Liu, Tian Yang, Yong Wang, Di Liu, Yuhong Jiang, Jie Huang, Jingwei Guo, Chunchao |
| author_facet | Zhao, Zibo Lai, Zeqiang Lin, Qingxiang Zhao, Yunfei Liu, Haolin Yang, Shuhui Feng, Yifei Yang, Mingxin Zhang, Sheng Yang, Xianghui Shi, Huiwen Liu, Sicong Wu, Junta Lian, Yihang Yang, Fan Tang, Ruining He, Zebin Wang, Xinzhou Liu, Jian Zuo, Xuhui Chen, Zhuo Lei, Biwen Weng, Haohan Xu, Jing Zhu, Yiling Liu, Xinhai Xu, Lixin Hu, Changrong Yang, Shaoxiong Zhang, Song Liu, Yang Huang, Tianyu Wang, Lifu Zhang, Jihong Chen, Meng Dong, Liang Jia, Yiwen Cai, Yulin Yu, Jiaao Tang, Yixuan Zhang, Hao Ye, Zheng He, Peng Wu, Runzhou Zhang, Chao Tan, Yonghao Xiao, Jie Tao, Yangyu Zhu, Jianchen Xue, Jinbao Liu, Kai Zhao, Chongqing Wu, Xinming Hu, Zhichao Qin, Lei Peng, Jianbing Li, Zhan Chen, Minghui Zhang, Xipeng Niu, Lin Wang, Paige Wang, Yingkai Kuang, Haozhao Fan, Zhongyi Zheng, Xu Zhuang, Weihao He, YingPing Liu, Tian Yang, Yong Wang, Di Liu, Yuhong Jiang, Jie Huang, Jingwei Guo, Chunchao |
| contents | We present Hunyuan3D 2.0, an advanced large-scale 3D synthesis system for generating high-resolution textured 3D assets. This system includes two foundation components: a large-scale shape generation model -- Hunyuan3D-DiT, and a large-scale texture synthesis model -- Hunyuan3D-Paint. The shape generative model, built on a scalable flow-based diffusion transformer, aims to create geometry that properly aligns with a given condition image, laying a solid foundation for downstream applications. The texture synthesis model, benefiting from strong geometric and diffusion priors, produces high-resolution and vibrant texture maps for either generated or hand-crafted meshes. Furthermore, we build Hunyuan3D-Studio -- a versatile, user-friendly production platform that simplifies the re-creation process of 3D assets. It allows both professional and amateur users to manipulate or even animate their meshes efficiently. We systematically evaluate our models, showing that Hunyuan3D 2.0 outperforms previous state-of-the-art models, including the open-source models and closed-source models in geometry details, condition alignment, texture quality, and etc. Hunyuan3D 2.0 is publicly released in order to fill the gaps in the open-source 3D community for large-scale foundation generative models. The code and pre-trained weights of our models are available at: https://github.com/Tencent/Hunyuan3D-2 |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_12202 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation Zhao, Zibo Lai, Zeqiang Lin, Qingxiang Zhao, Yunfei Liu, Haolin Yang, Shuhui Feng, Yifei Yang, Mingxin Zhang, Sheng Yang, Xianghui Shi, Huiwen Liu, Sicong Wu, Junta Lian, Yihang Yang, Fan Tang, Ruining He, Zebin Wang, Xinzhou Liu, Jian Zuo, Xuhui Chen, Zhuo Lei, Biwen Weng, Haohan Xu, Jing Zhu, Yiling Liu, Xinhai Xu, Lixin Hu, Changrong Yang, Shaoxiong Zhang, Song Liu, Yang Huang, Tianyu Wang, Lifu Zhang, Jihong Chen, Meng Dong, Liang Jia, Yiwen Cai, Yulin Yu, Jiaao Tang, Yixuan Zhang, Hao Ye, Zheng He, Peng Wu, Runzhou Zhang, Chao Tan, Yonghao Xiao, Jie Tao, Yangyu Zhu, Jianchen Xue, Jinbao Liu, Kai Zhao, Chongqing Wu, Xinming Hu, Zhichao Qin, Lei Peng, Jianbing Li, Zhan Chen, Minghui Zhang, Xipeng Niu, Lin Wang, Paige Wang, Yingkai Kuang, Haozhao Fan, Zhongyi Zheng, Xu Zhuang, Weihao He, YingPing Liu, Tian Yang, Yong Wang, Di Liu, Yuhong Jiang, Jie Huang, Jingwei Guo, Chunchao Computer Vision and Pattern Recognition We present Hunyuan3D 2.0, an advanced large-scale 3D synthesis system for generating high-resolution textured 3D assets. This system includes two foundation components: a large-scale shape generation model -- Hunyuan3D-DiT, and a large-scale texture synthesis model -- Hunyuan3D-Paint. The shape generative model, built on a scalable flow-based diffusion transformer, aims to create geometry that properly aligns with a given condition image, laying a solid foundation for downstream applications. The texture synthesis model, benefiting from strong geometric and diffusion priors, produces high-resolution and vibrant texture maps for either generated or hand-crafted meshes. Furthermore, we build Hunyuan3D-Studio -- a versatile, user-friendly production platform that simplifies the re-creation process of 3D assets. It allows both professional and amateur users to manipulate or even animate their meshes efficiently. We systematically evaluate our models, showing that Hunyuan3D 2.0 outperforms previous state-of-the-art models, including the open-source models and closed-source models in geometry details, condition alignment, texture quality, and etc. Hunyuan3D 2.0 is publicly released in order to fill the gaps in the open-source 3D community for large-scale foundation generative models. The code and pre-trained weights of our models are available at: https://github.com/Tencent/Hunyuan3D-2 |
| title | Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2501.12202 |