MambaIRv2: Attentive State Space Restoration
This work addresses image restoration challenges for computer vision applications, presenting an incremental improvement over existing Mamba-based methods.
The paper tackled the limitation of Mamba-based models in image restoration by proposing MambaIRv2, which introduces non-causal modeling and a semantic-guided mechanism, resulting in performance gains such as outperforming SRFormer by 0.35dB PSNR with fewer parameters and suppressing HAT by up to 0.29dB.
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.