UnfairGAN: An enhanced generative adversarial network for raindrop removal from a single image
作者：
Highlights：
• Developing an advanced loss function that can improve the instabilities of GAN.
• An advanced attention module can extract prior information to boost performance.
• A novel activation function can maximize information flow to pass through networks.
• Novel network architecture can probe hierarchical features for image restoration.
• A new large image dataset (DeepRaindrops) for training deraining models.
摘要
•Developing an advanced loss function that can improve the instabilities of GAN.•An advanced attention module can extract prior information to boost performance.•A novel activation function can maximize information flow to pass through networks.•Novel network architecture can probe hierarchical features for image restoration.•A new large image dataset (DeepRaindrops) for training deraining models.
论文关键词：Image deraining,Raindrop removal,Generative adversarial network,Deep Raindrops dataset
论文评审过程：Received 28 October 2021, Revised 21 June 2022, Accepted 17 July 2022, Available online 21 July 2022, Version of Record 17 August 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118232
