Frequency-Aware Adaptive Fusion Gate for Single Image Super-Resolution

Featured Application: The proposed Frequency-Aware Adaptive Fusion Gate (FAFG) for single image super-resolution holds significant potential for deployment in high-stakes professional domains that demand precise structural reconstruction, such as medical imaging diagnostics and public security. By ensuring the faithful recovery of complex geometric details while maintaining a plug-and-play architecture, our method accelerates the practical transition of advanced Transformer-based vision models to real-world industrial applications. Furthermore, its nearly zero-parameter frequency analysis design makes it highly viable for integration into resource-constrained edge devices, actively supporting highly resource-efficient and computationally friendly vision solutions. The Dense-Residual-Connected Transformer (DRCT) has established a new state-of-the-art in single image super-resolution by mitigating the information bottleneck in deep networks. However, its feature aggregation mechanism relies on a suboptimal Static Addition strategy, where residual features are scaled by a fixed, learnable scalar, regardless of the image content. This content-agnostic approach treats high-frequency textures and low-frequency noise indiscriminately, limiting the model’s representational capability. To address this, we propose a Frequency-Aware Adaptive Fusion Gate (FAFG) to replace the static scaling. Unlike spatial-only gating mechanisms, FAFG integrates the Discrete Cosine Transform (DCT) to explicitly perceive the frequency distribution of feature maps. By decomposing features into frequency components, our gate acts as an intelligent valve, dynamically amplifying valid structural details while suppressing redundant background noise. Extensive experiments on standard benchmarks demonstrate that our proposed FAFG-integrated model consistently outperforms the static-scaling and other state-of-the-art methods. Specifically, our method achieves a significant PSNR improvement of 0.31 dB on the texture-rich Urban100 dataset at (Formula presented.) scale. Visual results further confirm that our frequency-aware gating mechanism effectively recovers sharper edges and fine textures, providing a superior trade-off between reconstruction accuracy and model complexity.