Generative priors of large-scale text-to-image diffusion models enable a wide
range of new generation and editing applications on diverse visual modalities.
However, when adapting these priors to complex visual modalities, often
represented as multiple images (e.g., video), achieving consistency across a
set of images is challenging. In this paper, we address this challenge with a
novel method, Collaborative Score Distillation (CSD). CSD is based on the Stein
Variational Gradient Descent (SVGD). Specifically, we propose to consider
multiple samples as "particles" in the SVGD update and combine their score
functions to distill generative priors over a set of images synchronously.
Thus, CSD facilitates seamless integration of information across 2D images,
leading to a consistent visual synthesis across multiple samples. We show the
effectiveness of CSD in a variety of tasks, encompassing the visual editing of
panorama images, videos, and 3D scenes. Our results underline the competency of
CSD as a versatile method for enhancing inter-sample consistency, thereby
broadening the applicability of text-to-image diffusion models.

在多样化的视觉模态中，利用大规模文本生成模型进行生成和编辑应用时，跨多图像实现一致性是具有挑战性的，本文提出了一种基于 CSD 的新方法，该方法利用 SVGD 将多个样本视为 “粒子” 进行更新，并组合它们的评分函数以同步地蒸馏一组图像的生成先验，从而促进了 2D 图像信息的无缝集成，展示了 CSD 方法在全景图像、视频和 3D 场景等任务中的有效性，具有提高交叉样本一致性的实用性。