I came across conditional GAN, where the classes can be provided as input to the GAN. I would like to know, if the same network can be used for classes which can take any continuous value.ie, the generator once trained should be able to produce output for any condition, even if it was not a part of training set.
$\begingroup$
$\endgroup$
I think it exists as it is mentionned in the article Intra-class Variation Isolation in Conditional GANs from Richard T. Marriott, Sami Romdhani et Liming Chen in 5.2
Sadly I found nothing about those CC-GANs, eventhough it is said "classical"!