Recent advances in generative AI have unveiled significant potential for the creation of 3D content. However, current methods either apply a pre-trained 2D diffusion model with the time-consuming score distillation sampling (SDS), or a direct 3D diffusion model trained on limited 3D data losing generation diversity. In this work, we approach the problem by employing a multi-view 2.5D diffusion fine-tuned from a pre-trained 2D diffusion model. The multi-view 2.5D diffusion directly models the structural distribution of 3D data, while still maintaining the strong generalization ability of the original 2D diffusion model, filling the gap between 2D diffusion-based and direct 3D diffusion-based methods for 3D content generation. During inference, multi-view normal maps are generated using the 2.5D diffusion, and a novel differentiable rasterization scheme is introduced to fuse the almost consistent multi-view normal maps into a consistent 3D model. We further design a normal-conditioned multi-view image generation module for fast appearance generation given the 3D geometry. Our method is a one-pass diffusion process and does not require any SDS optimization as post-processing. We demonstrate through extensive experiments that, our direct 2.5D generation with the specially-designed fusion scheme can achieve diverse, mode-seeking-free, and high-fidelity 3D content generation in only 10 seconds. Project page: https://nju-3dv.github.io/projects/direct25.
@article{arxiv.2311.15980,
title = {Direct2.5: Diverse Text-to-3D Generation via Multi-view 2.5D Diffusion},
author = {Yuanxun Lu and Jingyang Zhang and Shiwei Li and Tian Fang and David McKinnon and Yanghai Tsin and Long Quan and Xun Cao and Yao Yao},
journal= {arXiv preprint arXiv:2311.15980},
year = {2024}
}
Comments
CVPR 2024 camera ready, including more evaluations and discussions. Project webpage: https://nju-3dv.github.io/projects/direct25