Adversarial Attack on Conditional Image Generative Models for Diverse and Controllable Generation

By introducing adversarial attack to pre-trained deterministic conditional image generative models, diverse and controllable results can be achieved without altering network structures or parameters.