Image based virtual try-on

• Main Author: Yong, Mark Wei Jie
• Other Authors: Lin Guosheng
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/165989
• Institution: Nanyang Technological University

Virtual try-ons have become an essential tool for online shopping, for providing customers with the option to try on clothing and other accessories in a virtual environment without needing to go down to the physical store. VITON-HD is a state-of-the-art virtual try-on system that uses high definition images and advanced deep learning algorithms to provide an accurate and realistic virtual try-on experience. This paper aims to dive into the methods of VITON-HD in virtual try-ons. VITON-HD can help address the challenges of virtual try-ons, such as inaccurate visual representations of clothing as well as incorrect fitting of clothing. This paper will review existing virtual try-on models, specifically VITON-HD about its preprocessing methods, experiments ran with VITON-HD as well as strengths and weakness of the models. Overall, this paper highlights the potential of VITON-HD in enhancing virtual try-ons and improving the online shopping experience for customers.