arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.04970cs.CV
收藏

LMR:一个用于基于参考的超分辨率的大规模多参考数据集

LMR: A Large-Scale Multi-Reference Dataset for Reference-based Super-Resolution

Baidu Inc.(百度公司);CASIA(中国科学院自动化研究所)

Lin Zhang, Xin Li, Dongliang He, Errui Ding, Zhaoxiang Zhang

摘要:

人们普遍认为,与单图像超分辨率(SISR)相比,基于参考的超分辨率(RefSR)通过参考类似的高质量图像来实现更优越的结果。直觉上,引用越多,性能越好。然而,以前的RefSR方法都集中在单参考图像训练上,而在测试或实际应用中通常可以使用多个参考图像。这种训练-测试不匹配的根本原因是缺乏公开可用的多参考SR训练数据集,这极大地阻碍了多参考超分辨率的研究工作。为此,我们构建了一个大规模,多参考超分辨率数据集,命名为LMR。它包含112,142组300 x300训练图像,是现有最大RefSR数据集的10倍。图像尺寸也大得多。更重要的是,每个组配备有5个具有不同相似度的参考图像。此外,我们提出了一种新的基线方法多参考超分辨率:MRefSR,包括一个多参考注意模块(MAM)的特征融合的任意数量的参考图像,和一个空间感知滤波模块(SAFM)的融合特征选择。拟议的MRefSR实现了显着的改进,国家的最先进的方法,定量和定性评价。我们的代码和数据将很快提供。

英文摘要:

It is widely agreed that reference-based super-resolution (RefSR) achieves superior results by referring to similar high quality images, compared to single image super-resolution (SISR). Intuitively, the more references, the better performance. However, previous RefSR methods have all focused on single-reference image training, while multiple reference images are often available in testing or practical applications. The root cause of such training-testing mismatch is the absence of publicly available multi-reference SR training datasets, which greatly hinders research efforts on multi-reference super-resolution. To this end, we construct a large-scale, multi-reference super-resolution dataset, named LMR. It contains 112,142 groups of 300x300 training images, which is 10x of the existing largest RefSR dataset. The image size is also much larger. More importantly, each group is equipped with 5 reference images with different similarity levels. Furthermore, we propose a new baseline method for multi-reference super-resolution: MRefSR, including a Multi-Reference Attention Module (MAM) for feature fusion of an arbitrary number of reference images, and a Spatial Aware Filtering Module (SAFM) for the fused feature selection. The proposed MRefSR achieves significant improvements over state-of-the-art approaches on both quantitative and qualitative evaluations. Our code and data would be made available soon.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑