GRA-GAN: Generative adversarial network for image style transfer of Gender, Race, and age
作者：
Highlights：
• Generating gender, age, and race images based on information fusion in GRA-GAN.
• Element-wise multiplication of the image gradient is performed in the decoder.
• Unidirectional training was configured for forward and backward directions.
• For training, fusing the race and gender tasks, and configuring reference age loss.
• Our trained model are publicly available.
摘要
•Generating gender, age, and race images based on information fusion in GRA-GAN.•Element-wise multiplication of the image gradient is performed in the decoder.•Unidirectional training was configured for forward and backward directions.•For training, fusing the race and gender tasks, and configuring reference age loss.•Our trained model are publicly available.
论文关键词：Facial image transformation,Age estimation and classification of race and gender,GRA-GAN,Channel-wise and multiplication-based information fusion of encoder and decoder features
论文评审过程：Received 4 November 2021, Revised 25 January 2022, Accepted 27 February 2022, Available online 8 March 2022, Version of Record 11 March 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.116792
