There are two losses used in Stage-I of StackGAN as follows:
- Generator loss
- Discriminator loss
The discriminator loss can be represented as follows:
The preceding equation is pretty self-explanatory. It represents the loss function for the discriminator network, in which both networks are conditioned on the text embeddings.
The generator loss can be represented as follows:
The preceding equation is pretty self-explanatory too. It represents the loss function for the generator network, in which both networks are conditioned on the text embeddings. Also, it includes a KL-divergence term to the loss function.