Fashion-AttGAN: Attribute-Aware Fashion Editing With Multi-Objective GAN

摘要

In this paper, we introduce a novel task, namely attribute-aware fashion-editing, to the fashion domain. A first dataset is constructed for this task with 14,221 images and 22 attributes, which has been made publically available. We re-define the overall objectives in AttGAN and propose the Fashion-AttGAN model to tackle this new task. Experimental results show effectiveness of our Fashion-AttGAN on fashion editing over the original AttGAN.