Boundary Equilibrium Generative Adversarial Networks

Loss Function

• using l1 loss. You can see the details in began_model.py line 233~236.

Architecture Networks

• Maybe, same as in the BEGAN paper. But *different hyper-parameters :).

Tensorboard

Result

Name	Global Step 50k	Global Step 100k	Global Step 300k
BEGAN

Took about 28800 seconds on GTX 1080.

To-Do

• Add Loss Function & Explains