Generative Adversarial Network

A type of deep learning architecture consisting of two neural networks — a generator and a discriminator — that are trained in competition with each other. The generator creates synthetic data (such as images) while the discriminator tries to distinguish between real and generated data. Through this adversarial process, the generator learns to produce increasingly realistic outputs. In accessibility, GANs have been applied to tasks including generating multi-view lip visualizations for speech-reading assistance, creating synthetic training data for sign language recognition, enhancing images for people with low vision, and producing realistic avatars for sign language animation.