Generating a Fusion Image: One's Identity and Another's Shape
This addresses image manipulation for applications like pose transfer or style synthesis, but it is incremental as it builds on existing GAN frameworks.
The paper tackles the problem of generating a fusion image that combines the identity of one input image with the shape of another, using a GAN-based network with identity and shape losses and a Min-Patch training method. It demonstrates qualitative results on multiple datasets including VGG Youtube Pose, Eye, and Photo-Sketch-Cartoon.
Generating a novel image by manipulating two input images is an interesting research problem in the study of generative adversarial networks (GANs). We propose a new GAN-based network that generates a fusion image with the identity of input image x and the shape of input image y. Our network can simultaneously train on more than two image datasets in an unsupervised manner. We define an identity loss LI to catch the identity of image x and a shape loss LS to get the shape of y. In addition, we propose a novel training method called Min-Patch training to focus the generator on crucial parts of an image, rather than its entirety. We show qualitative results on the VGG Youtube Pose dataset, Eye dataset (MPIIGaze and UnityEyes), and the Photo-Sketch-Cartoon dataset.