FreqMamba：从频率视角审视用于图像去雨的 Mamba
FreqMamba: Viewing Mamba from a Frequency Perspective for Image Deraining
AI总结：
本文提出 FreqMamba，将 Mamba 与频带交互、傅里叶全局建模及多尺度雨天退化先验结合，用于图像去雨，并在视觉与定量结果上超过现有方法。
摘要：
被雨纹破坏的图像通常会丢失对感知至关重要的频率信息，而图像去雨旨在解决这一问题，该任务依赖于全局和局部退化建模。近期研究见证了 Mamba 在感知全局与局部信息方面的有效性和高效性，其基础是利用图像块之间的局部相关性；然而，很少有研究尝试将其与频率分析相结合并扩展到图像去雨任务，这限制了其感知与频率建模（例如 Fourier transform，傅里叶变换）相关的全局退化的能力。本文提出 FreqMamba，这是一种有效且高效的范式，利用 Mamba 与频率分析之间的互补性进行图像去雨。该方法的核心在于从两个视角用频率分析扩展 Mamba：通过频带进行扩展以利用频率相关性，并将其与 Fourier transform 相连接以进行全局退化建模。具体而言，FreqMamba 引入了互补的三重交互结构，包括 spatial Mamba、frequency band Mamba 和 Fourier global modeling。frequency band Mamba 将图像分解为不同频率的子带，从而能够从频率维度进行二维扫描。此外，利用 Mamba 独特的数据依赖特性，我们使用不同尺度的雨天图像为网络提供退化先验，从而促进高效训练。大量实验表明，本方法在视觉效果和定量指标上均优于最先进方法。
英文摘要：
Images corrupted by rain streaks often lose vital frequency information for perception, and image deraining aims to solve this issue which relies on global and local degradation modeling. Recent studies have witnessed the effectiveness and efficiency of Mamba for perceiving global and local information based on its exploiting local correlation among patches, however, rarely attempts have been explored to extend it with frequency analysis for image deraining, limiting its ability to perceive global degradation that is relevant to frequency modeling (e.g. Fourier transform). In this paper, we propose FreqMamba, an effective and efficient paradigm that leverages the complementary between Mamba and frequency analysis for image deraining. The core of our method lies in extending Mamba with frequency analysis from two perspectives: extending it with frequency-band for exploiting frequency correlation, and connecting it with Fourier transform for global degradation modeling. Specifically, FreqMamba introduces complementary triple interaction structures including spatial Mamba, frequency band Mamba, and Fourier global modeling. Frequency band Mamba decomposes the image into sub-bands of different frequencies to allow 2D scanning from the frequency dimension. Furthermore, leveraging Mamba's unique data-dependent properties, we use rainy images at different scales to provide degradation priors to the network, thereby facilitating efficient training. Extensive experiments show that our method outperforms state-of-the-art methods both visually and quantitatively.
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。
