Dress-up: Deep Neural Framework for Image-based Human Appearance Transfer

作     者：Ghodhbani, Hajer Neji, Mohamed Qahtani, Abdulrahman M. Almutiry, Omar Dhahri, Habib Alimi, Adel M. 

作者机构： BP 1173 Sfax3038 Tunisia National School of Electronics and Telecommunications of Sfax Technopark BP 1163 SfaxCP 3018 Tunisia Department of Computer Science College of Computers and Information Technology Taif University P.O.Box. 11099 Taif21944 Saudi Arabia College of Applied Computer Science King Saud University Riyadh Saudi Arabia Department of Electrical and Electronic Engineering Science Faculty of Engineering and the Built Environment University of Johannesburg South Africa 

出 版 物：《TechRxiv》 (TechRxiv)

年 卷 期：2022年

核心收录：

主　　题：Textures 

摘      要：The fashion industry is at the brink of radical transformation. The emergence of Artificial Intelligence (AI) in fashion applications creates many opportunities for this industry. Interesting to this matter, we proposed a flexible person generation system for virtual try-on, presented in this paper, and aiming to treat the task of human appearance transfer across images while preserving texture details and structural coherence of the generated outfit. This challenging task has drawn increasing attention and made huge development of intelligent fashion applications. However, this task requires different challenges, especially, in the case of a wide divergences between the source and target images. To solve this problem, we proposed a flexible person generation framework called Dress-up to treat the 2D virtual try-on task. Dress-up is an end-to-end generation pipeline with three modules based on the task of image-to-image translation aiming to sequentially interchange garments between images, and produce dressing effects not achievable by existing works. The core idea of our solution is to explicitly encode the body pose and the target clothes by a pre-processing module based on the segmentation process. Then, a conditional adversarial network is implemented to generate target segmentation feeding to the alignment and translation networks to generate the final output. Our experimental results on the widely used DeepFashion dataset demonstrate that Dress-up provides significant results over the state-of-the-art methods on the subject of output quality despite different interactions of garments, and handles the virtual try-on task with a weakly supervised manner. © 2022, CC BY.