提出PS-Mamba，让状态空间模型更好恢复图像细节与全局结构。
Progressive Split Mamba: Effective State Space Modelling for Image Restoration
- 分层分区处理图像，保持局部邻域关系
- 多尺度跳跃连接缓解远距离信息衰减
- 适合高保真图像修复任务，优于现有模型
图像修复需同时保留精细局部结构并维持长程空间一致性。尽管卷积网络受限于感受野，而Transformer在全局注意力上存在二次复杂度问题，近期的状态空间模型（SSM）如Mamba为长程依赖建模提供了线性时间的替代方案。然而，将Mamba直接扩展到2D图像会暴露两个固有缺陷：首先，将2D特征图展平为1D序列破坏了空间拓扑，导致局部失真，影响结构恢复精度；其次，基于稳定性的递归动态使信息随空间距离呈指数衰减，削弱全局一致性。为此，我们提出渐进式分块状态空间模型（PS-Mamba），一种拓扑感知的分层状态空间框架，兼顾局部保持与高效全局传播。不同于整体展平，PS-Mamba进行几何一致的分块处理，保持邻域完整性；通过逐级分裂（二分、四分、八分）实现结构化多尺度建模，同时保持线性复杂度。为抑制长程衰减，引入对称跨尺度捷径路径，直接传递低频全局上下文，稳定大范围信息流动。在超分辨率、去噪和JPEG压缩伪影去除任务上的大量实验表明，该方法持续优于近期的Mamba及注意力基模型，提升明显。
原文摘要 · Abstract (English)
Image restoration requires simultaneously preserving fine-grained local structures and maintaining long-range spatial coherence. While convolutional networks struggle with limited receptive fields, and Transformers incur quadratic complexity for global attention, recent State Space Models (SSMs), such as Mamba, provide an appealing linear-time alternative for long-range dependency modelling. However, naively extending Mamba to 2D images exposes two intrinsic shortcomings. First, flattening 2D feature maps into 1D sequences disrupts spatial topology, leading to locality distortion that hampers precise structural recovery. Second, the stability-driven recurrent dynamics of SSMs induce long-range decay, progressively attenuating information across distant spatial positions and weakening global consistency. Together, these effects limit the effectiveness of state-space modelling in high-fidelity restoration. We propose Progressive Split-Mamba (PS-Mamba), a topology-aware hierarchical state-space framework designed to reconcile locality preservation with efficient global propagation. Instead of sequentially flattening entire feature maps, PS-Mamba performs geometry-consistent partitioning, maintaining neighbourhood integrity prior to state-space processing. A progressive split hierarchy (halves, quadrants, octants) enables structured multi-scale modelling while retaining linear complexity. To counteract long-range decay, we introduce symmetric cross-scale shortcut pathways that directly transmit low-frequency global context across hierarchical levels, stabilising information flow over large spatial extents. Extensive experiments on super-resolution, denoising, and JPEG artifact reduction show consistent improvements over recent Mamba-based and attention-based models with a clear margin.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。