Asymmetric Cyclegan for Unpaired NIR-to-RGB Face Image Translation
Hao Dou, Chen Chen, Xiyuan Hu, Silong Peng
Abstract
Translating near-infrared (NIR) face into color (RGB) face, is helpful to improve the visual effect of images and the performance of face recognition. The model for unpaired image-to-image translation is suitable for this task due to the high cost of pixel-matched data. Because of the complexity difference between NIR and RGB image domains, the complexity inequality in bidirectional NIR-RGB translations is significant. We analyze the limitation of the original CycleGAN in asymmetric translation tasks, and propose an Asymmetric Cycle-GAN model with U-net-like generators of unequal sizes to adapt to the asymmetric need in NIR-RGB translations. The edge-retain loss between NIR and the generated RGB images is also introduced to enhance face visual quality. The qualitative visual evaluation and quantitative evaluation with face ID and skin color criteria show that our model achieves great improvements compared with state-of-the-art methods on three public datasets and a newly proposed dataset.
BibTeX
@inproceedings{icassp2019_asymmetriccycleg,
title = {Asymmetric Cyclegan for Unpaired NIR-to-RGB Face Image Translation},
author = {Hao Dou and Chen Chen and Xiyuan Hu and Silong Peng},
booktitle = {ICASSP 2019},
year = {2019}
}