Saved in:
Bibliographic Details
Main Authors: Wang, Yanwen, Zhuang, Yiyu, Zhang, Jiawei, Wang, Li, Zeng, Yifei, Cao, Xun, Zuo, Xinxin, Zhu, Hao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.02466
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • In this paper, we rethink text-to-avatar generative models by proposing TeRA, a more efficient and effective framework than the previous SDS-based models and general large 3D generative models. Our approach employs a two-stage training strategy for learning a native 3D avatar generative model. Initially, we distill a decoder to derive a structured latent space from a large human reconstruction model. Subsequently, a text-controlled latent diffusion model is trained to generate photorealistic 3D human avatars within this latent space. TeRA enhances the model performance by eliminating slow iterative optimization and enables text-based partial customization through a structured 3D human representation. Experiments have proven our approach's superiority over previous text-to-avatar generative models in subjective and objective evaluation.