Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution

doi:10.1109/TIP.2017.2750403

CORC > 北京大学 > 信息科学技术学院

	Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution
	Yang, Wenhan ; Feng, Jiashi ; Yang, Jianchao ; Zhao, Fang ; Liu, Jiaying ; Guo, Zongming ; Yan, Shuicheng
刊名	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2017
关键词	Super-resolution edge guidance recurrent residual network sub-band recovery LINEAR INVERSE PROBLEMS QUALITY ASSESSMENT REGULARIZATION NETWORK
DOI	10.1109/TIP.2017.2750403
英文摘要	In this paper, we consider the image superresolution (SR) problem. The main challenge of image SR is to recover high-frequency details of a low-resolution (LR) image that are important for human perception. To address this essentially ill-posed problem, we introduce a Deep Edge Guided REcurrent rEsidual (DEGREE) network to progressively recover the high-frequency details. Different from most of the existing methods that aim at predicting high-resolution (HR) images directly, the DEGREE investigates an alternative route to recover the difference between a pair of LR and HR images by recurrent residual learning. DEGREE further augments the SR process with edge-preserving capability, namely the LR image and its edge map can jointly infer the sharp edge details of the HR image during the recurrent recovery process. To speed up its training convergence rate, by-pass connections across the multiple layers of DEGREE are constructed. In addition, we offer an understanding on DEGREE from the view-point of sub-band frequency decomposition on image signal and experimentally demonstrate how the DEGREE can recover different frequency bands separately. Extensive experiments on three benchmark data sets clearly demonstrate the superiority of DEGREE over the well-established baselines and DEGREE also provides new state-of-the-arts on these data sets. We also present addition experiments for JPEG artifacts reduction to demonstrate the good generality and flexibility of our proposed DEGREE network to handle other image processing tasks.; National Natural Science Foundation of China [61772043]; CCF-Tencent Open Research Fund; SCI(E); ARTICLE; 12; 5895-5907; 26
语种	英语
内容类型	期刊论文
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/470357]
专题	信息科学技术学院
推荐引用方式 GB/T 7714	Yang, Wenhan,Feng, Jiashi,Yang, Jianchao,et al. Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2017.
APA	Yang, Wenhan.,Feng, Jiashi.,Yang, Jianchao.,Zhao, Fang.,Liu, Jiaying.,...&Yan, Shuicheng.(2017).Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution.IEEE TRANSACTIONS ON IMAGE PROCESSING.
MLA	Yang, Wenhan,et al."Deep Edge Guided Recurrent Residual Learning for Image Super-Resolution".IEEE TRANSACTIONS ON IMAGE PROCESSING (2017).