Score: 0

TeRA: Rethinking Text-guided Realistic 3D Avatar Generation

Published: September 2, 2025 | arXiv ID: 2509.02466v1

By: Yanwen Wang , Yiyu Zhuang , Jiawei Zhang and more

Potential Business Impact:

Creates realistic 3D people from text descriptions.

Business Areas:
3D Technology Hardware, Software

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.

Page Count
17 pages

Category
Computer Science:
CV and Pattern Recognition