Super-Resolution via Conditional Implicit Maximum Likelihood Estimation

Single-image super-resolution (SISR) is a canonical problem with diverse applications. Leading methods like SRGAN produce images that contain various artifacts, such as high-frequency noise, hallucinated colours and shape distortions, which adversely affect the realism of the result. In this paper, we propose an alternative approach based on an extension of the method of Implicit Maximum Likelihood Estimation (IMLE). We demonstrate greater effectiveness at noise reduction and preservation of the original colours and shapes, yielding more realistic super-resolved images.

[1]  Yingyu Liang,et al.  Generalization and Equilibrium in Generative Adversarial Nets (GANs) , 2017, ICML.

[2]  Joan Bruna,et al.  Intriguing properties of neural networks , 2013, ICLR.

[3]  Ian J. Goodfellow,et al.  On distinguishability criteria for estimating generative models , 2014, ICLR.

[4]  Thomas B. Moeslund,et al.  Super-resolution: a comprehensive survey , 2014, Machine Vision and Applications.

[5]  Jitendra Malik,et al.  Fast k-Nearest Neighbour Search via Prioritized DCI , 2017, ICML.

[6]  Jitendra Malik,et al.  Fast k-Nearest Neighbour Search via Dynamic Continuous Indexing , 2015, ICML.

[7]  Xiaoou Tang,et al.  Image Super-Resolution Using Deep Convolutional Networks , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Ritabrata Dutta,et al.  Likelihood-free inference via classification , 2014, Stat. Comput..

[9]  Joan Bruna,et al.  Super-Resolution with Deep Convolutional Sufficient Statistics , 2015, ICLR.

[10]  Lei Zhang,et al.  Convolutional Sparse Coding for Image Super-Resolution , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[11]  Jitendra Malik,et al.  Implicit Maximum Likelihood Estimation , 2018, ArXiv.

[12]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[13]  Michael Elad,et al.  Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries , 2006, IEEE Transactions on Image Processing.

[14]  C. Duchon.  Lanczos Filtering in One and Two Dimensions , 1979 .

[15]  Michal Irani,et al.  Super-resolution from a single image , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[16]  William T. Freeman,et al.  Example-Based Super-Resolution , 2002, IEEE Computer Graphics and Applications.

[17]  Thomas S. Huang,et al.  Image Super-Resolution Via Sparse Representation , 2010, IEEE Transactions on Image Processing.

[18]  Christian Ledig,et al.  Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  Chih-Yuan Yang,et al.  Single-Image Super-Resolution: A Benchmark , 2014, ECCV.

[20]  Raanan Fattal,et al.  Image and video upscaling from local self-examples , 2011, TOGS.

[21]  Yi Zhang,et al.  Do GANs actually learn the distribution? An empirical study , 2017, ArXiv.

[22]  Narendra Ahuja,et al.  Single image super-resolution from transformed self-exemplars , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).