HY3D-Bench: Generation of 3D Assets
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917246878613504 |
|---|---|
| author | Hunyuan3D, Team : Zhang, Bowen Guo, Chunchao Guo, Dongyuan Liu, Haolin Yan, Hongyu Shi, Huiwen Yu, Jiaao Xu, Jiachen Huang, Jingwei Li, Kunhong Wang, Lifu Linus Wang, Penghao Lin, Qingxiang Tang, Ruining Yang, Xianghui Li, Yang Guan, Yirui Zhao, Yunfei Yang, Yunhan Lai, Zeqiang Liang, Zhihao Zhao, Zibo |
| author_facet | Hunyuan3D, Team : Zhang, Bowen Guo, Chunchao Guo, Dongyuan Liu, Haolin Yan, Hongyu Shi, Huiwen Yu, Jiaao Xu, Jiachen Huang, Jingwei Li, Kunhong Wang, Lifu Linus Wang, Penghao Lin, Qingxiang Tang, Ruining Yang, Xianghui Li, Yang Guan, Yirui Zhao, Yunfei Yang, Yunhan Lai, Zeqiang Liang, Zhihao Zhao, Zibo |
| contents | While recent advances in neural representations and generative models have revolutionized 3D content creation, the field remains constrained by significant data processing bottlenecks. To address this, we introduce HY3D-Bench, an open-source ecosystem designed to establish a unified, high-quality foundation for 3D generation. Our contributions are threefold: (1) We curate a library of 250k high-fidelity 3D objects distilled from large-scale repositories, employing a rigorous pipeline to deliver training-ready artifacts, including watertight meshes and multi-view renderings; (2) We introduce structured part-level decomposition, providing the granularity essential for fine-grained perception and controllable editing; and (3) We bridge real-world distribution gaps via a scalable AIGC synthesis pipeline, contributing 125k synthetic assets to enhance diversity in long-tail categories. Validated empirically through the training of Hunyuan3D-2.1-Small, HY3D-Bench democratizes access to robust data resources, aiming to catalyze innovation across 3D perception, robotics, and digital content creation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03907 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HY3D-Bench: Generation of 3D Assets Hunyuan3D, Team : Zhang, Bowen Guo, Chunchao Guo, Dongyuan Liu, Haolin Yan, Hongyu Shi, Huiwen Yu, Jiaao Xu, Jiachen Huang, Jingwei Li, Kunhong Wang, Lifu Linus Wang, Penghao Lin, Qingxiang Tang, Ruining Yang, Xianghui Li, Yang Guan, Yirui Zhao, Yunfei Yang, Yunhan Lai, Zeqiang Liang, Zhihao Zhao, Zibo Computer Vision and Pattern Recognition Artificial Intelligence While recent advances in neural representations and generative models have revolutionized 3D content creation, the field remains constrained by significant data processing bottlenecks. To address this, we introduce HY3D-Bench, an open-source ecosystem designed to establish a unified, high-quality foundation for 3D generation. Our contributions are threefold: (1) We curate a library of 250k high-fidelity 3D objects distilled from large-scale repositories, employing a rigorous pipeline to deliver training-ready artifacts, including watertight meshes and multi-view renderings; (2) We introduce structured part-level decomposition, providing the granularity essential for fine-grained perception and controllable editing; and (3) We bridge real-world distribution gaps via a scalable AIGC synthesis pipeline, contributing 125k synthetic assets to enhance diversity in long-tail categories. Validated empirically through the training of Hunyuan3D-2.1-Small, HY3D-Bench democratizes access to robust data resources, aiming to catalyze innovation across 3D perception, robotics, and digital content creation. |
| title | HY3D-Bench: Generation of 3D Assets |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2602.03907 |