Generating a Fusion Image: One's Identity and Another's Shape

TL;DR

This paper introduces a GAN-based method for creating fusion images that combine the identity of one image with the shape of another, trained unsupervised across multiple datasets.

Contribution

A novel GAN architecture with identity and shape loss functions, and a new Min-Patch training method for focused image generation.

Findings

Effective fusion of identity and shape in generated images

Unsupervised training across multiple datasets

Focus on crucial image parts improves quality

Abstract

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.