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 2206.00941cs.LGcs.AIcs.CVstat.ML
收藏

利用流形约束改进用于逆问题的扩散模型

Improving Diffusion Models for Inverse Problems using Manifold Constraints

Korea Advanced Institute of Science and Technology (KAIST)(韩国科学技术院);Kim Jaechul Graduate School of AI(金在哲人工智能研究生院)

Hyungjin Chung, Byeongsu Sim, Dohoon Ryu, Jong Chul Ye

AI总结:

针对现有扩散模型逆问题求解器易偏离数据流形、误差累积的问题,提出基于流形约束的校正项,可与原有方法协同提升性能,在多类图像逆任务中效果更优。

摘要:

近年来,扩散模型已被用于以无监督方式求解各类逆问题,只需对采样过程进行适当修改即可。然而,现有求解器会递归执行反向扩散步骤,随后再执行基于投影的测量一致性步骤,往往会产生次优结果。通过研究生成采样路径,本文指出现有求解器会使采样路径偏离数据流形,进而导致误差累积。为解决这一问题,我们受流形约束启发提出了一个额外的校正项,它可与现有求解器协同作用,使迭代过程贴近流形。所提出的流形约束仅需几行代码即可轻松实现,却能大幅提升性能。通过大量实验,我们证明该方法在理论和实证上均优于现有方法,在图像修复、图像上色、稀疏视角计算机断层扫描等多种应用中取得了可观的效果。代码可在https://github.com/HJ-harry/MCG_diffusion获取。

英文摘要:

Recently, diffusion models have been used to solve various inverse problems in an unsupervised manner with appropriate modifications to the sampling process. However, the current solvers, which recursively apply a reverse diffusion step followed by a projection-based measurement consistency step, often produce suboptimal results. By studying the generative sampling path, here we show that current solvers throw the sample path off the data manifold, and hence the error accumulates. To address this, we propose an additional correction term inspired by the manifold constraint, which can be used synergistically with the previous solvers to make the iterations close to the manifold. The proposed manifold constraint is straightforward to implement within a few lines of code, yet boosts the performance by a surprisingly large margin. With extensive experiments, we show that our method is superior to the previous methods both theoretically and empirically, producing promising results in many applications such as image inpainting, colorization, and sparse-view computed tomography. Code available https://github.com/HJ-harry/MCG_diffusion

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。

↑