arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2206.04029cs.CVcs.LG
收藏

加速高分辨率图像合成的基于得分的生成模型

Accelerating Score-based Generative Models for High-Resolution Image Synthesis

Fudan University(复旦大学);University of Surrey(萨里大学);RIKEN(理化学研究所)

Hengyuan Ma, Li Zhang, Xiatian Zhu, Jingfeng Zhang, Jianfeng Feng

摘要:

基于得分的生成模型是近年来新兴的一类生成模型。其关键思想是通过向高斯样本反复添加高斯噪声和梯度,直到收敛到目标分布,即,扩散采样。然而,为了确保采样收敛的稳定性和生成质量,该顺序采样过程必须采用小的步长和许多采样迭代(例如,2000年)。已经提出了几种加速方法,重点是低分辨率生成。在这项工作中,我们认为加速与SGM的高分辨率生成,一个更具挑战性,但更重要的问题。我们从理论上证明,这种缓慢的收敛缺点主要是由于目标分布的无知。此外,我们引入了一种新的目标分布感知采样(TDAS)方法,利用空间和频率域的结构先验。在CIFAR-10、CelebA、LSUN和FFHQ数据集上进行的大量实验验证了TDAS可以始终如一地加速最先进的SGM,特别是在更具挑战性的高分辨率(1024 x1024)图像生成任务上,速度高达18.4倍,同时在很大程度上保持了合成质量。通过较少的采样迭代,TDAS仍然可以生成高质量的图像。相比之下,现有的方法性能急剧下降,甚至完全失效

英文摘要:

Score-based generative models (SGMs) have recently emerged as a promising class of generative models. The key idea is to produce high-quality images by recurrently adding Gaussian noises and gradients to a Gaussian sample until converging to the target distribution, a.k.a. the diffusion sampling. To ensure stability of convergence in sampling and generation quality, however, this sequential sampling process has to take a small step size and many sampling iterations (e.g., 2000). Several acceleration methods have been proposed with focus on low-resolution generation. In this work, we consider the acceleration of high-resolution generation with SGMs, a more challenging yet more important problem. We prove theoretically that this slow convergence drawback is primarily due to the ignorance of the target distribution. Further, we introduce a novel Target Distribution Aware Sampling (TDAS) method by leveraging the structural priors in space and frequency domains. Extensive experiments on CIFAR-10, CelebA, LSUN, and FFHQ datasets validate that TDAS can consistently accelerate state-of-the-art SGMs, particularly on more challenging high resolution (1024x1024) image generation tasks by up to 18.4x, whilst largely maintaining the synthesis quality. With fewer sampling iterations, TDAS can still generate good quality images. In contrast, the existing methods degrade drastically or even fails completely

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑