Super-Resolution via Conditional Implicit Maximum Likelihood Estimation [article]

Ke Li, Shichong Peng, Jitendra Malik
2018 arXiv   pre-print
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
more » ... original colours and shapes, yielding more realistic super-resolved images.
arXiv:1810.01406v1 fatcat:5ew3exgylja55jfnuidgnpxzbu