Text-to-3D modelling has seen exciting progress by combining generative text-to-image models with image-to-3D methods like Neural Radiance Fields. DreamFusion recently achieved high-quality results but requires a lengthy, per-prompt optimization to create 3D objects. To address this, we amortize optimization over text prompts by training on many prompts simultaneously with a unified model, instead of separately. With this, we share computation across a prompt set, training in less time than per-prompt optimization. Our framework - Amortized text-to-3D (ATT3D) - enables knowledge-sharing between prompts to generalize to unseen setups and smooth interpolations between text for novel assets and simple animations.
@article{arxiv.2306.07349,
title = {ATT3D: Amortized Text-to-3D Object Synthesis},
author = {Jonathan Lorraine and Kevin Xie and Xiaohui Zeng and Chen-Hsuan Lin and Towaki Takikawa and Nicholas Sharp and Tsung-Yi Lin and Ming-Yu Liu and Sanja Fidler and James Lucas},
journal= {arXiv preprint arXiv:2306.07349},
year = {2023}
}