阅读论文-Flow-based Kernel Prior with Application to Blind Super-Resolution Shiyu Paper reading/super-resolution 发布于:Jul 6, 2021 FKP(flow-based kernel prior);出处:CVPR 2021 阅读全文
阅读论文-Unsupervised Degradation Representation Learning for Blind Super-Resolution Shiyu Paper reading/super-resolution 发布于:Jul 5, 2021 DASR (degradation-aware SR);无监督对比学习得到degradation representation;出处:CVPR 2021 阅读全文
阅读论文-Invertible Image Rescaling Shiyu Paper reading/super-resolution 发布于:Jul 4, 2021 Invertible rescaling;出处:ECCV 2020 oral 阅读全文
阅读论文-Learning Continuous Image Representation with Local Implicit Image Function Shiyu Paper reading/super-resolution 发布于:Jul 2, 2021 LIIF;出处:CVPR 2021 oral 阅读全文
阅读论文-Analyzing and Mitigating Compression Defects in Deep Learning Shiyu Paper reading/LV+HV 发布于:Jul 1, 2021 compression artifacts correction + segmentation/classification/detection;出处:arxiv 2020.11 阅读全文
阅读论文-Deep Plug-and-Play Super-Resolution for Arbitrary Blur Kernels Shiyu Paper reading/super-resolution 发布于:Jun 24, 2021 plug-and-play;出处:CVPR2019 阅读全文
阅读论文-Image Hallucination with Primal Sketch Priors Shiyu Paper reading/super-resolution 发布于:Jun 16, 2021 sketch priors, CVPR 2003 paper motivation 不依赖于prior training 在待测试的图像上训练,利用图像内部的冗余信息。 在acquisition process未知的图像上超过已有的SOTA supervised 的CNN和unsupervised方法。 “supervised” for any method that trains on externally supplied examples (even if their generation does not require manuallabelling). contribution 设计了 Method
阅读论文-“Zero-Shot” Super-Resolution using Deep Internal Learning Shiyu Paper reading/super-resolution 发布于:Jun 4, 2021 ZSSR;出处:CVPR2018 阅读全文