Face Manipulation with Generative Adversarial Network

Abstract Generative adversarial network has appeared as an effective image manipulation tool in recent years and has been widely used. The GAN-based manipulation of face images is also possible and tools including DeepFake are already misused. In this paper, we discuss the pros and cons of face manipulation with generative adversarial network. We find that this technique can be very useful for recovering masked face and further improving face recognition accuracy.