Transformer-Progressive Mamba Network for Lightweight Image Super-Resolution
This work addresses a specific bottleneck in lightweight image super-resolution for applications requiring efficient high-quality upscaling, representing an incremental improvement over existing Mamba-based approaches.
The paper tackles the problem of limited fine-grained transitions across modeling scales in Mamba-based super-resolution methods by proposing T-PMambaSR, a lightweight framework that integrates window-based self-attention with Progressive Mamba and an Adaptive High-Frequency Refinement Module, resulting in better performance and lower computational cost than recent Transformer- or Mamba-based methods.
Recently, Mamba-based super-resolution (SR) methods have demonstrated the ability to capture global receptive fields with linear complexity, addressing the quadratic computational cost of Transformer-based SR approaches. However, existing Mamba-based methods lack fine-grained transitions across different modeling scales, which limits the efficiency of feature representation. In this paper, we propose T-PMambaSR, a lightweight SR framework that integrates window-based self-attention with Progressive Mamba. By enabling interactions among receptive fields of different scales, our method establishes a fine-grained modeling paradigm that progressively enhances feature representation with linear complexity. Furthermore, we introduce an Adaptive High-Frequency Refinement Module (AHFRM) to recover high-frequency details lost during Transformer and Mamba processing. Extensive experiments demonstrate that T-PMambaSR progressively enhances the model's receptive field and expressiveness, yielding better performance than recent Transformer- or Mamba-based methods while incurring lower computational cost. Our codes will be released after acceptance.