HiFA: High-fidelity Text-to-3D Generation with Advanced Diffusion Guidance

arXiv:2305.18766v4 Announce Type: replace-cross
Abstract: The advancements in automatic text-to-3D generation have been remarkable. Most existing methods use pre-trained text-to-image diffusion models to optimize 3D representations like Neural Radiance Fields (NeRFs) via latent-space denoising score matching. Yet, these methods often result in artifacts and inconsistencies across different views due to their suboptimal optimization approaches and limited understanding of 3D geometry. Moreover, the inherent constraints of NeRFs in rendering crisp geometry and stable textures usually lead to a two-stage optimization to …

abstract advanced arxiv cs.ai cs.cv cs.lg denoising diffusion diffusion models fidelity fields guidance image image diffusion latent-space neural radiance fields space text text-to-image type via