CU-Mamba: Selective State Space Models with Channel Learning for Image Restoration
This work addresses image restoration for computer vision applications, presenting a novel hybrid method that is incremental but with specific architectural innovations.
The paper tackles image reconstruction by proposing CU-Mamba, a model that integrates spatial and channel state space models into a U-Net architecture to address limitations like poor long-range dependency modeling and high computational costs in existing methods. It demonstrates superior performance over state-of-the-art approaches in experiments.
Reconstructing degraded images is a critical task in image processing. Although CNN and Transformer-based models are prevalent in this field, they exhibit inherent limitations, such as inadequate long-range dependency modeling and high computational costs. To overcome these issues, we introduce the Channel-Aware U-Shaped Mamba (CU-Mamba) model, which incorporates a dual State Space Model (SSM) framework into the U-Net architecture. CU-Mamba employs a Spatial SSM module for global context encoding and a Channel SSM component to preserve channel correlation features, both in linear computational complexity relative to the feature map size. Extensive experimental results validate CU-Mamba's superiority over existing state-of-the-art methods, underscoring the importance of integrating both spatial and channel contexts in image restoration.