用轻量设计提升卷积网络超分辨率速度，实现实时推理。
PlainUSR: Chasing Faster ConvNet for Efficient Super-Resolution
- 通过重参数化将MobileNetv3转为更快的普通卷积块。
- 引入局部重要性注意力，在1阶延迟内实现高阶信息交互。
- 采用纯通道分裂拼接的U-Net主干，兼顾速度与性能。
降低延迟是近期超分辨率（SR）研究的主流趋势。尽管已有工作通过各类卷积模块、注意力机制和主干网络挖掘卷积神经网络（ConvNet）潜力，但实现实时性能仍具挑战。为此，本文提出PlainUSR，通过三项关键改进加速ConvNet用于高效超分辨率。在卷积块层面，利用重参数化技巧将较轻但较慢的MobileNetv3块转化为更重但更快的普通卷积；在注意力模块中，通过区域重要性图与门控调制输入，引入基于局部重要性的注意力，实现高阶信息交互，仅需1阶注意力延迟；在主干网络方面，提出纯化的U-Net结构，执行通道级判别性分割与拼接。实验表明，PlainUSR展现出极低延迟、良好可扩展性及与当前顶尖速度与质量导向方法相当的性能。尤其相比最新NGswin，PlainUSR-L提速达16.4倍，同时保持竞争力。
原文摘要 · Abstract (English)
Reducing latency is a roaring trend in recent super-resolution (SR) research. While recent progress exploits various convolutional blocks, attention modules, and backbones to unlock the full potentials of the convolutional neural network (ConvNet), achieving real-time performance remains a challenge. To this end, we present PlainUSR, a novel framework incorporating three pertinent modifications to expedite ConvNet for efficient SR. For the convolutional block, we squeeze the lighter but slower MobileNetv3 block into a heavier but faster vanilla convolution by reparameterization tricks to balance memory access and calculations. For the attention module, by modulating input with a regional importance map and gate, we introduce local importance-based attention to realize high-order information interaction within a 1-order attention latency. As to the backbone, we propose a plain U-Net that executes channel-wise discriminate splitting and concatenation. In the experimental phase, PlainUSR exhibits impressively low latency, great scalability, and competitive performance compared to both state-of-the-art latency-oriented and quality-oriented methods. In particular, compared to recent NGswin, the PlainUSR-L is 16.4x faster with competitive performance.
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。