2020
Self-Supervised CycleGAN for Object-Preserving Image-to-Image Domain Adaptation
ECCV 2020poster
Recent generative adversarial network (GAN) based methods (e.g., CycleGAN) are prone to fail at preserving image-objects in image-to-image translation, which reduces their practicality on tasks such as domain adaptation. Some frameworks have been proposed to adopt a segmentation network as the auxil…