报告人：汪朝辉 巴黎东大学
报告时间：7月23日 10:00-11:00
地点：1208
摘要：
This talk will exhibits two of our recent works in deep modeling for object image re-rendering and domain adaptation: (i) a principled Tag Disentangled Generative Adversarial Networks (TD-GAN) for re-rendering new images for the object of inter- est from a single image of it by specifying multiple scene properties (such as viewpoint, illumination, expression, etc.); (2) a geometry-consistent generative adversarial network (Gc-GAN), which enables one-sided unsupervised domain mapping.
