Researched
Generative Adversarial Networks
Goodfellow's GAN (2014) trains a generator and a discriminator against each other and produces the first photorealistic fake images.
Open in the interactive tree →The two-player training idea gave sharp synthetic faces, images and later video. GANs were the leading image generators until diffusion models took over.
Prerequisites
- Deep Learning2012
Unlocks
- Generative Image & Video AI2022GANs gave the first photorealistic generated images