How to control the class to which the image generated by the GAN model generator belongs?

Deep Learning
1

The main idea of GAN-like models is ’ the game of trying to generate simulation images and trying to distinguish between true and false image models.’

My point of doubt is: in this process, how to control the class to which the “model used to generate simulation image” generated image belongs?
Or does it mean that in the GAN-like model, it is impossible to determine which class the generated simulation image belongs to?

2

This is work going into that direction:


(A super in-depth tutorial you can find here: https://www.depthfirstlearning.com/2018/InfoGAN)

1 Like
3

Thanks for your help. :grinning: