ClawGAN: Claw connection-based generative adversarial networks for facial image translation in thermal to RGB visible light

作者：

Highlights：

• ClawGAN for Thermal-Visible facial image translation is proposed.

• We propose the Mismatch metric (MM) to measure mapping relationship of paired images.

• We introduce the generative reconstructed loss and the synthesized loss to full objective.

• The claw connection network architecture are used as the generator of GAN.

摘要

•ClawGAN for Thermal-Visible facial image translation is proposed.•We propose the Mismatch metric (MM) to measure mapping relationship of paired images.•We introduce the generative reconstructed loss and the synthesized loss to full objective.•The claw connection network architecture are used as the generator of GAN.

论文关键词：Image translation,Mismatch metric,Claw connections,Generating adversarial networks,Thermal facial images

论文评审过程：Received 13 April 2021, Revised 16 August 2021, Accepted 20 November 2021, Available online 28 November 2021, Version of Record 3 December 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116269