GANs: This AI Algorithm Works Like A Game - How do Generative Adversarial Networks Work?

preview_player
Показать описание
One of the most interesting machine learning algorithms are Generative Adversarial Networks or GANs for short. They have been widely used in generating realistic images and videos and have applications in domains, such as art, fashion, and entertainment. They consist of two neural networks: a generator and a discriminator, engaged in a competitive game.

Video footage from: Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold, by Pan, Xingang and Tewari, Ayush, and Leimkühler, Thomas and Liu, Lingjie and Meka, Abhimitra and Theobalt, Christian (ACM SIGGRAPH 2023 Conference Proceedings):

Join my channel membership to support my work:

#generativeai #ai #aitools #neuralnetworks #gan #draggan
Рекомендации по теме
Комментарии
Автор

To avoid mode collapse shouldn't you just add more real data?

Democracy_Manifest
Автор

What is the software used in the video?

SlfgjkAldfjgf
Автор

why does this video have so little comments xD

keyknot