MambaIRv2：注意力状态空间图像复原
MambaIRv2: Attentive State Space Restoration
AI总结：
本文提出 MambaIRv2，通过注意力状态空间方程和语义引导邻近机制赋予 Mamba 非因果建模能力，在轻量级与经典超分辨率任务上以更少参数超越 SRFormer 和 HAT。
摘要：
基于 Mamba 的图像复原骨干网络近期在平衡全局感受野与计算效率方面展现出显著潜力。然而，Mamba 固有的因果建模局限——每个 token 仅依赖于扫描序列中其之前的元素——限制了图像中像素信息的充分利用，从而为图像复原带来了新的挑战。本文提出 MambaIRv2，使 Mamba 具备类似于 ViT 的非因果建模能力，以实现注意力状态空间复原模型。具体而言，所提出的注意力状态空间方程能够关注扫描序列之外的区域，并仅通过一次扫描即可实现图像展开。此外，我们进一步引入语义引导的邻近机制，以促进远距离但相似的像素之间的交互。大量实验表明，MambaIRv2 在轻量级超分辨率任务上比 SRFormer 的 PSNR 高出 0.35dB，且参数量减少 9.3%；在经典超分辨率任务上比 HAT 最高提升 0.29dB。代码开源在 https://github.com/csguoh/MambaIR。
英文摘要：
The Mamba-based image restoration backbones have recently demonstrated significant potential in balancing global reception and computational efficiency. However, the inherent causal modeling limitation of Mamba, where each token depends solely on its predecessors in the scanned sequence, restricts the full utilization of pixels across the image and thus presents new challenges in image restoration. In this work, we propose MambaIRv2, which equips Mamba with the non-causal modeling ability similar to ViTs to reach the attentive state space restoration model. Specifically, the proposed attentive state-space equation allows to attend beyond the scanned sequence and facilitate image unfolding with just one single scan. Moreover, we further introduce a semantic-guided neighboring mechanism to encourage interaction between distant but similar pixels. Extensive experiments show our MambaIRv2 outperforms SRFormer by even 0.35dB PSNR for lightweight SR even with 9.3\% less parameters and suppresses HAT on classic SR by up to 0.29dB. Code is available at https://github.com/csguoh/MambaIR.
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。
