How it works...

The generator learns how to produce forged images of celebrities and the discriminator learns how to distinguish the forged images from the real ones. Each epoch in the two networks competes to improve and reduce losses. The first five epochs are reported in the following table:

Epoch 0

Epoch 1

Epoch 2

Epoch 3

Epoch 4

Epoch 5

Example of forged celebrities with a DCGAN 
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.143.115.131