BriefGPT.xyz
Jun, 2024
文本到图像矫正流作为即插即用的先验
Text-to-Image Rectified Flow as Plug-and-Play Priors
HTML
PDF
Xiaofeng Yang, Cheng Chen, Xulei Yang, Fayao Liu, Guosheng Lin
TL;DR
大规模扩散模型在生成任务中取得了显著的性能。矫正流是一种新的生成模型类别,在各个领域展现出了优越的性能。通过理论和实验证据,我们展示了基于矫正流的方法与扩散模型具有相似的功能,可以作为有效的先验。此外,基于矫正流的先验方法在图像反转方面表现出优越的性能。
Abstract
large-scale diffusion models
have achieved remarkable performance in
generative tasks
. Beyond their initial training applications, these models have proven their ability to function as versatile plug-and-play pri
→