arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.04747cs.CVcs.AIcs.LG
收藏

视觉中的扩散模型:一项综述

Diffusion Models in Vision: A Survey

University of Bucharest(布加勒斯特大学);University of Central Florida(中佛罗里达大学);Center for Research in Computer Vision (CRCV)(计算机视觉研究中心(CRCV))

Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu, Mubarak Shah

AI总结:

该综述全面梳理了视觉领域去噪扩散模型的理论与实践进展,介绍三类通用框架、对比其与其他深度生成模型的关系,完成多视角分类并指出局限与未来方向。

摘要:

去噪扩散模型是计算机视觉领域近期新兴的研究方向,在生成建模领域展现出了卓越的成果。扩散模型是一种深度生成模型,包含两个阶段:前向扩散阶段和反向扩散阶段。在前向扩散阶段,输入数据通过添加高斯噪声,在多个步骤中被逐渐扰动。在反向阶段,模型的任务是通过学习逐步逆转扩散过程,一步步恢复原始输入数据。扩散模型因生成样本的质量和多样性而广受认可,尽管它们存在已知的计算负担,即由于采样过程涉及大量步骤而导致速度较慢。在本综述中,我们对应用于视觉领域的去噪扩散模型相关论文进行了全面梳理,涵盖该领域的理论与实践贡献。首先,我们梳理并介绍了三种通用扩散建模框架,分别基于去噪扩散概率模型(denoising diffusion probabilistic models)、噪声条件分数网络(noise conditioned score networks)和随机微分方程(stochastic differential equations)。我们进一步讨论了扩散模型与其他深度生成模型之间的关系,包括变分自编码器、生成对抗网络、基于能量的模型、自回归模型和归一化流。随后,我们对应用于计算机视觉的扩散模型进行了多视角的分类。最后,我们阐明了扩散模型当前的局限性,并展望了未来研究的一些有趣方向。

英文摘要:

Denoising diffusion models represent a recent emerging topic in computer vision, demonstrating remarkable results in the area of generative modeling. A diffusion model is a deep generative model that is based on two stages, a forward diffusion stage and a reverse diffusion stage. In the forward diffusion stage, the input data is gradually perturbed over several steps by adding Gaussian noise. In the reverse stage, a model is tasked at recovering the original input data by learning to gradually reverse the diffusion process, step by step. Diffusion models are widely appreciated for the quality and diversity of the generated samples, despite their known computational burdens, i.e. low speeds due to the high number of steps involved during sampling. In this survey, we provide a comprehensive review of articles on denoising diffusion models applied in vision, comprising both theoretical and practical contributions in the field. First, we identify and present three generic diffusion modeling frameworks, which are based on denoising diffusion probabilistic models, noise conditioned score networks, and stochastic differential equations. We further discuss the relations between diffusion models and other deep generative models, including variational auto-encoders, generative adversarial networks, energy-based models, autoregressive models and normalizing flows. Then, we introduce a multi-perspective categorization of diffusion models applied in computer vision. Finally, we illustrate the current limitations of diffusion models and envision some interesting directions for future research.

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。

↑