LucidDreamer：通过区间得分匹配实现高保真度文本到3D生成

LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching

November 19, 2023

作者: Yixun Liang, Xin Yang, Jiantao Lin, Haodong Li, Xiaogang Xu, Yingcong Chen

cs.AI

摘要

最近在文本到3D生成领域的进展标志着生成模型中的重要里程碑，为在各种现实场景中创造富有想象力的3D资产开启了新的可能性。虽然最近在文本到3D生成方面取得了一些进展，但往往在渲染详细和高质量的3D模型方面表现不佳。这个问题特别普遍，因为许多方法基于得分蒸馏采样（SDS）。本文指出了SDS存在的一个显著缺陷，即为3D模型带来不一致和低质量的更新方向，导致过度平滑效果。为了解决这个问题，我们提出了一种名为区间得分匹配（ISM）的新方法。ISM采用确定性扩散轨迹，并利用基于区间的得分匹配来抵消过度平滑。此外，我们将3D高斯光斑投影技术纳入我们的文本到3D生成流程中。大量实验证明，我们的模型在质量和训练效率方面大大优于现有技术水平。

English

The recent advancements in text-to-3D generation mark a significant milestone in generative models, unlocking new possibilities for creating imaginative 3D assets across various real-world scenarios. While recent advancements in text-to-3D generation have shown promise, they often fall short in rendering detailed and high-quality 3D models. This problem is especially prevalent as many methods base themselves on Score Distillation Sampling (SDS). This paper identifies a notable deficiency in SDS, that it brings inconsistent and low-quality updating direction for the 3D model, causing the over-smoothing effect. To address this, we propose a novel approach called Interval Score Matching (ISM). ISM employs deterministic diffusing trajectories and utilizes interval-based score matching to counteract over-smoothing. Furthermore, we incorporate 3D Gaussian Splatting into our text-to-3D generation pipeline. Extensive experiments show that our model largely outperforms the state-of-the-art in quality and training efficiency.