A Deep Dual-Branch Networks for Joint Blind Motion Deblurring and Super-Resolution

被引:2
作者
Zhang, Xinyi [1 ]
Wang, Fei [1 ]
Dong, Hang [1 ]
Guo, Yu [1 ]
机构
[1] Xi An Jiao Tong Univ, Inst Artificial Intelligence & Robot, 28 Xianning West Rd, Xian 710049, Shaanxi, Peoples R China
来源
PROCEEDINGS OF THE 2ND INTERNATIONAL CONFERENCE ON VISION, IMAGE AND SIGNAL PROCESSING (ICVISP 2018) | 2018年
关键词
Super-resolution; Blind Motion Deblurring; Dual Network; RESOLUTION;
D O I
10.1145/3271553.3271554
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Image super-resolution is a fundamental pre-processing technique for the machine vision applications of robotics and other mobile platforms. Inevitably, images captured by the mobile camera tend to emerge severe motion blur and this degradation will deteriorate the performance of current state-of-the-art super-resolution methods. In this paper, we propose a deep dual-branch convolution neural network (CNN) to generate a clear high-resolution image from a single natural image with severe blurs. Compared to off-the-shelf methods, our method, called DB-SRN, can remove the complex non-uniform motion blurs and restore useful texture details simultaneously. By sharing the features from modified residual blocks (ResBlocks), the dual-branch design can promote the performances of both tasks other while retaining network simplicity. Extensive experiments demonstrate that our method produces remarkable deblurred and super-resolved images in terms of quality and quantity with high computational efficiency.
引用
收藏
页数:6
相关论文
共 17 条
[1]  
[Anonymous], ARXIV161202177
[2]  
[Anonymous], 2017, ARXIV170306870
[3]   Evaluation of image resolution and super-resolution on face recognition performance [J].
Fookes, Clinton ;
Lin, Frank ;
Chandran, Vinod ;
Sridharan, Sridha .
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2012, 23 (01) :75-93
[4]   From Motion Blur to Motion Flow: a Deep Learning Solution for Removing Heterogeneous Motion Blur [J].
Gong, Dong ;
Yang, Jie ;
Liu, Lingqiao ;
Zhang, Yanning ;
Reid, Ian ;
Shen, Chunhua ;
van den Hengel, Anton ;
Shi, Qinfeng .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :3806-3815
[5]  
HRADIS M, 2015, P BMVC, V10
[6]   IMPROVING RESOLUTION BY IMAGE REGISTRATION [J].
IRANI, M ;
PELEG, S .
CVGIP-GRAPHICAL MODELS AND IMAGE PROCESSING, 1991, 53 (03) :231-239
[7]  
Kim J, 2016, PROC CVPR IEEE, P1637, DOI [10.1109/CVPR.2016.182, 10.1109/CVPR.2016.181]
[8]   Recording and Playback of Camera Shake: Benchmarking Blind Deconvolution with a Real-World Database [J].
Koehler, Rolf ;
Hirsch, Michael ;
Mohler, Betty ;
Schoelkopf, Bernhard ;
Harmeling, Stefan .
COMPUTER VISION - ECCV 2012, PT VII, 2012, 7578 :27-40
[9]   Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution [J].
Lai, Wei-Sheng ;
Huang, Jia-Bin ;
Ahuja, Narendra ;
Yang, Ming-Hsuan .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :5835-5843
[10]   Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network [J].
Ledig, Christian ;
Theis, Lucas ;
Huszar, Ferenc ;
Caballero, Jose ;
Cunningham, Andrew ;
Acosta, Alejandro ;
Aitken, Andrew ;
Tejani, Alykhan ;
Totz, Johannes ;
Wang, Zehan ;
Shi, Wenzhe .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :105-114