Diverse adversarial network for image super-resolution

作者：

Highlights：

• We propose a new variation to generative adversarial networks to address the well-known problems of image super resolution.

• We proposed a model to increase the variation in generated sample.

• Our proposed model employs multi generators instead of single generators.

• We adopted least square loss function for handling the diverse samples.

摘要

•We propose a new variation to generative adversarial networks to address the well-known problems of image super resolution.•We proposed a model to increase the variation in generated sample.•Our proposed model employs multi generators instead of single generators.•We adopted least square loss function for handling the diverse samples.

论文关键词：Super-resolution,Adversarial network,Diverse GAN,Deep learning

论文评审过程：Received 23 October 2018, Revised 16 February 2019, Accepted 18 February 2019, Available online 6 March 2019, Version of Record 9 March 2019.

论文官网地址：https://doi.org/10.1016/j.image.2019.02.008