Due to the recent progress made by state-of-the-art deep learning approaches, the development of facial age transformation algorithms has become an attractive research topic in the fields of computer vision. Given a face as input, the facial age transformation refers to the generation of facial images for the input face but with older or younger ages in the sense that the identity of the input face can be well preserved. This is a challenging task due to the intrinsic complexity of the facial appearance variation caused by the physical aging process, which can be related to individual physical condition, gender, race and other actors. It has received increasing attention in recent years because of the effectiveness of the Generative Adversarial Network (GAN) based approaches, the availability of large facial age datasets and commercial potentials.

AgeTransGAN consists of an encoder-decoder generator G = [Gen, Gde], where Gen is the encoder and Gde is the decoder, and a conditional multitask discriminator Dp. 

➢  The generator G = [Gen, Gde] with networks Bid and Bag for age and identity disentanglement. 

➢  The Conditional Multitask Projection (CMP) discriminator Dp with four subnets for multilayer feature extraction and an age classifier Ca.