HumanGAN: A Generative Model of Humans Images

Abstract

Generative adversarial networks achieve great performance in photorealistic image synthesis in various domains, including human images. However, they usually employ latent vector inputs encoding sampled output images globally. This does not allow convenient control of semantically-relevant individual parts of the image, and is not able to draw samples that only differ in partial aspects, such asclothing style. We address these limitations and present a generative model for images of dressed humans offering control over pose, local body part appearance and garment style. This is the first method to solve various aspect of human image generation such as global appearance sampling, pose transfer, parts and garment transfer, and parts ampling jointly in an unified framework. As our model encodes part-based latent appearance vectors in a normalized pose-independent space and warps them to different poses, it preserves body and clothing appearance under varying posture. Experiments show that our flexible and general generative method outperforms task-specific baselines for pose-conditioned image generation, pose transfer and part sampling in terms of realism and output resolution.