用分层结构提升Mamba模型图像超分辨率效率，减少扫描次数。
Hi-Mamba: Hierarchical Mamba for Efficient Image Super-Resolution
- 分层模块结合局部与区域状态空间模型，单向扫描建模多尺度上下文。
- 在×3超分辨率任务中，相比MambaIR提升0.29 dB PSNR，速度更快。
- 适合追求高效高精度图像重建的视觉算法研究者使用。
状态空间模型（SSM）如Mamba在建模长程依赖方面表现出色，具有线性复杂度，在从高层到低层视觉任务中均取得成功应用。然而，由于其序列特性，需在不同方向进行多次扫描以弥补图像展开为一维序列后空间依赖性的损失，这显著增加了计算开销，难以应用于高分辨率图像处理。为此，我们提出一种新的分层Mamba网络——Hi-Mamba，用于图像超分辨率（SR）。Hi-Mamba包含两个关键设计：(1) 分层状态空间模块（HMB），由局部SSM（L-SSM）和区域SSM（R-SSM）组成，均采用单向扫描，聚合多尺度表示以增强上下文建模能力；(2) 方向交替分层组（DA-HMG），将同构的单向扫描分配至级联的HMB中，以丰富空间关系建模。大量实验表明，Hi-Mamba在五个基准数据集上均表现出优越性能。例如，在×3超分辨率任务中，相对于强大的轻量级MambaIR，Hi-Mamba在Manga109数据集上实现0.29 dB的显著PSNR提升。
原文摘要 · Abstract (English)
State Space Models (SSM), such as Mamba, have shown strong representation ability in modeling long-range dependency with linear complexity, achieving successful applications from high-level to low-level vision tasks. However, SSM's sequential nature necessitates multiple scans in different directions to compensate for the loss of spatial dependency when unfolding the image into a 1D sequence. This multi-direction scanning strategy significantly increases the computation overhead and is unbearable for high-resolution image processing. To address this problem, we propose a novel Hierarchical Mamba network, namely, Hi-Mamba, for image super-resolution (SR). Hi-Mamba consists of two key designs: (1) The Hierarchical Mamba Block (HMB) assembled by a Local SSM (L-SSM) and a Region SSM (R-SSM) both with the single-direction scanning, aggregates multi-scale representations to enhance the context modeling ability. (2) The Direction Alternation Hierarchical Mamba Group (DA-HMG) allocates the isomeric single-direction scanning into cascading HMBs to enrich the spatial relationship modeling. Extensive experiments demonstrate the superiority of Hi-Mamba across five benchmark datasets for efficient SR. For example, Hi-Mamba achieves a significant PSNR improvement of 0.29 dB on Manga109 for $\times3$ SR, compared to the strong lightweight MambaIR.
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。