轻量级超分网络Sebica，用双向注意力提升清晰度，参数少至3%仍保高画质。
Sebica: Lightweight Spatial and Efficient Bidirectional Channel Attention Super Resolution Network
- 引入空间与双向通道注意力机制，降低计算开销。
- 在Div2K/Flickr2K上达28.29/0.7976和30.18/0.8330的性能表现。
- 模型仅需3%参数与算力，适合移动端或实时场景部署。
单图像超分辨率（SISR）是提升低分辨率图像视觉质量的关键技术。尽管深度学习模型在该领域取得显著进展，但常面临计算资源消耗大的问题，限制了其在资源受限或实时性要求高的环境中的应用。为此，我们提出Sebica，一种融合空间与高效双向通道注意力机制的轻量级网络。Sebica在大幅降低计算成本的同时保持高重建质量，在Div2K和Flickr2K数据集上的PSNR/SSIM分别为28.29/0.7976和30.18/0.8330，超越多数轻量级基线模型，并接近最优模型表现，但仅需其17%和15%的参数量与GFLOPs。其小型版本仅含7.9K参数与0.41 GFLOPs，为最优模型的3%，在Flickr2K上仍实现28.12/0.7931的PSNR/SSIM，且在交通视频等真实场景中显著提升目标检测准确率。
原文摘要 · Abstract (English)
Single Image Super-Resolution (SISR) is a vital technique for improving the visual quality of low-resolution images. While recent deep learning models have made significant advancements in SISR, they often encounter computational challenges that hinder their deployment in resource-limited or time-sensitive environments. To overcome these issues, we present Sebica, a lightweight network that incorporates spatial and efficient bidirectional channel attention mechanisms. Sebica significantly reduces computational costs while maintaining high reconstruction quality, achieving PSNR/SSIM scores of 28.29/0.7976 and 30.18/0.8330 on the Div2K and Flickr2K datasets, respectively. These results surpass most baseline lightweight models and are comparable to the highest-performing model, but with only 17% and 15% of the parameters and GFLOPs. Additionally, our small version of Sebica has only 7.9K parameters and 0.41 GFLOPS, representing just 3% of the parameters and GFLOPs of the highest-performing model, while still achieving PSNR and SSIM metrics of 28.12/0.7931 and 0.3009/0.8317, on the Flickr2K dataset respectively. In addition, Sebica demonstrates significant improvements in real-world applications, specifically in object detection tasks, where it enhances detection accuracy in traffic video scenarios.
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。