轻量级超分模型UCAN融合卷积与注意力，高效扩展感受野。
UCAN: Unified Convolutional Attention Network for Expansive Receptive Fields in Lightweight Super-Resolution
- 统一卷积与注意力机制，结合窗口注意力与刺猬注意力建模长程依赖。
- 48.4G MACs下在Manga109上达31.63 dB PSNR，超越同类轻量模型。
- 适合资源受限设备部署，兼顾精度、效率与可扩展性。
混合CNN-Transformer架构在图像超分辨率中表现优异，但扩大注意力窗口或卷积核会显著增加计算成本，限制在资源受限设备上的部署。本文提出UCAN，一种轻量级网络，通过统一卷积与注意力机制高效扩展有效感受野。UCAN结合基于窗口的空间注意力与刺猬注意力（Hedgehog Attention）以建模局部纹理与长程依赖，并引入基于知识蒸馏的大核模块，在不增加过多计算量的前提下保留高频结构。此外，采用跨层参数共享进一步降低复杂度。在Manga109（4×）上，UCAN-L仅需48.4G MACs即达31.63 dB PSNR，优于近期轻量级模型；在BSDS100上达27.79 dB，超越参数量更大的方法。大量实验表明，UCAN在精度、效率与可扩展性间实现更优权衡，适用于实际高分辨率图像修复任务。
原文摘要 · Abstract (English)
Hybrid CNN-Transformer architectures achieve strong results in image super-resolution, but scaling attention windows or convolution kernels significantly increases computational cost, limiting deployment on resource-constrained devices. We present UCAN, a lightweight network that unifies convolution and attention to expand the effective receptive field efficiently. UCAN combines window-based spatial attention with a Hedgehog Attention mechanism to model both local texture and long-range dependencies, and introduces a distillation-based large-kernel module to preserve high-frequency structure without heavy computation. In addition, we employ cross-layer parameter sharing to further reduce complexity. On Manga109 ($4\times$), UCAN-L achieves 31.63 dB PSNR with only 48.4G MACs, surpassing recent lightweight models. On BSDS100, UCAN attains 27.79 dB, outperforming methods with significantly larger models. Extensive experiments show that UCAN achieves a superior trade-off between accuracy, efficiency, and scalability, making it well-suited for practical high-resolution image restoration.
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。