Identity-aware CycleGAN for face photo-sketch synthesis and recognition

作者：

Highlights：

• Jointly solve synthesis and recognition problem by their close relationship.

• Identity-Aware CycleGAN (IACycleGAN) model with a perceptual loss in CycleGAN to supervise the image generation network.

• A mutual optimization procedure between the synthesis model and recognition model.

• Extensive experiments are performed on both photo-to-sketch and sketch-to-photo tasks using the widely used CUFS and CUFSF databases.

摘要

•Jointly solve synthesis and recognition problem by their close relationship.•Identity-Aware CycleGAN (IACycleGAN) model with a perceptual loss in CycleGAN to supervise the image generation network.•A mutual optimization procedure between the synthesis model and recognition model.•Extensive experiments are performed on both photo-to-sketch and sketch-to-photo tasks using the widely used CUFS and CUFSF databases.

论文关键词：Convolutional neural network,Generative adversarial network,Photo-sketch synthesis,Photo-sketch recognition,Identity-aware training

论文评审过程：Received 18 June 2019, Revised 11 December 2019, Accepted 26 January 2020, Available online 27 January 2020, Version of Record 8 February 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107249