A Comparative Study of Image-to-Image Translation Using GANs for Synthetic Child Race Data
This addresses data scarcity for children in face recognition, but is incremental as it applies existing methods to a new domain.
The study tackled the lack of ethnic diversity in child face data by using image-to-image translation with GANs to synthesize Caucasian and Asian child data, demonstrating feasibility through experimental validation.
The lack of ethnic diversity in data has been a limiting factor of face recognition techniques in the literature. This is particularly the case for children where data samples are scarce and presents a challenge when seeking to adapt machine vision algorithms that are trained on adult data to work on children. This work proposes the utilization of image-to-image transformation to synthesize data of different races and thus adjust the ethnicity of children's face data. We consider ethnicity as a style and compare three different Image-to-Image neural network based methods, specifically pix2pix, CycleGAN, and CUT networks to implement Caucasian child data and Asian child data conversion. Experimental validation results on synthetic data demonstrate the feasibility of using image-to-image transformation methods to generate various synthetic child data samples with broader ethnic diversity.