Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lai, Zeqiang, Zhao, Yunfei, Liu, Haolin, Zhao, Zibo, Lin, Qingxiang, Shi, Huiwen, Yang, Xianghui, Yang, Mingxin, Yang, Shuhui, Feng, Yifei, Zhang, Sheng, Huang, Xin, Luo, Di, Yang, Fan, Yang, Fang, Wang, Lifu, Liu, Sicong, Tang, Yixuan, Cai, Yulin, He, Zebin, Liu, Tian, Liu, Yuhong, Jiang, Jie, Linus, Huang, Jingwei, Guo, Chunchao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913903812804608
author Lai, Zeqiang
Zhao, Yunfei
Liu, Haolin
Zhao, Zibo
Lin, Qingxiang
Shi, Huiwen
Yang, Xianghui
Yang, Mingxin
Yang, Shuhui
Feng, Yifei
Zhang, Sheng
Huang, Xin
Luo, Di
Yang, Fan
Yang, Fang
Wang, Lifu
Liu, Sicong
Tang, Yixuan
Cai, Yulin
He, Zebin
Liu, Tian
Liu, Yuhong
Jiang, Jie
Linus
Huang, Jingwei
Guo, Chunchao
author_facet Lai, Zeqiang
Zhao, Yunfei
Liu, Haolin
Zhao, Zibo
Lin, Qingxiang
Shi, Huiwen
Yang, Xianghui
Yang, Mingxin
Yang, Shuhui
Feng, Yifei
Zhang, Sheng
Huang, Xin
Luo, Di
Yang, Fan
Yang, Fang
Wang, Lifu
Liu, Sicong
Tang, Yixuan
Cai, Yulin
He, Zebin
Liu, Tian
Liu, Yuhong
Jiang, Jie
Linus
Huang, Jingwei
Guo, Chunchao
contents In this report, we present Hunyuan3D 2.5, a robust suite of 3D diffusion models aimed at generating high-fidelity and detailed textured 3D assets. Hunyuan3D 2.5 follows two-stages pipeline of its previous version Hunyuan3D 2.0, while demonstrating substantial advancements in both shape and texture generation. In terms of shape generation, we introduce a new shape foundation model -- LATTICE, which is trained with scaled high-quality datasets, model-size, and compute. Our largest model reaches 10B parameters and generates sharp and detailed 3D shape with precise image-3D following while keeping mesh surface clean and smooth, significantly closing the gap between generated and handcrafted 3D shapes. In terms of texture generation, it is upgraded with phyiscal-based rendering (PBR) via a novel multi-view architecture extended from Hunyuan3D 2.0 Paint model. Our extensive evaluation shows that Hunyuan3D 2.5 significantly outperforms previous methods in both shape and end-to-end texture generation.
format Preprint
id arxiv_https___arxiv_org_abs_2506_16504
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details
Lai, Zeqiang
Zhao, Yunfei
Liu, Haolin
Zhao, Zibo
Lin, Qingxiang
Shi, Huiwen
Yang, Xianghui
Yang, Mingxin
Yang, Shuhui
Feng, Yifei
Zhang, Sheng
Huang, Xin
Luo, Di
Yang, Fan
Yang, Fang
Wang, Lifu
Liu, Sicong
Tang, Yixuan
Cai, Yulin
He, Zebin
Liu, Tian
Liu, Yuhong
Jiang, Jie
Linus
Huang, Jingwei
Guo, Chunchao
Computer Vision and Pattern Recognition
Artificial Intelligence
In this report, we present Hunyuan3D 2.5, a robust suite of 3D diffusion models aimed at generating high-fidelity and detailed textured 3D assets. Hunyuan3D 2.5 follows two-stages pipeline of its previous version Hunyuan3D 2.0, while demonstrating substantial advancements in both shape and texture generation. In terms of shape generation, we introduce a new shape foundation model -- LATTICE, which is trained with scaled high-quality datasets, model-size, and compute. Our largest model reaches 10B parameters and generates sharp and detailed 3D shape with precise image-3D following while keeping mesh surface clean and smooth, significantly closing the gap between generated and handcrafted 3D shapes. In terms of texture generation, it is upgraded with phyiscal-based rendering (PBR) via a novel multi-view architecture extended from Hunyuan3D 2.0 Paint model. Our extensive evaluation shows that Hunyuan3D 2.5 significantly outperforms previous methods in both shape and end-to-end texture generation.
title Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2506.16504