Machine learning for digital try-on: Challenges and progress
- PDF / 1,549,788 Bytes
- 9 Pages / 612 x 808 pts Page_size
- 16 Downloads / 218 Views
Machine learning for digital try-on: Challenges and progress Junbang Liang1 (
), Ming C. Lin1
c The Author(s) 2020.
while the ultimate goal is a fast and realistic try-on system adaptive to each customer’s body. There is still a substantial technological gap between modeling and demonstrating garment fitting in the digital and real worlds, including fast and realistic demonstration, accurate modeling of human body and garments, faithful modeling of garment material, and lossless transformation of garments between virtual and physical worlds. In this paper, we present some open research issues Keywords machine learning; digital try-on; garment that contribute to this technological gap, including: modeling; human body estimation; material 1. accurate estimation of human shapes and sizes modeling using consumer devices, 2. faithful recovery of garment materials via (online) images, and 1 Introduction 3. ease of design and manipulation of sewing patterns E-commerce has grown at a rapid pace in recent and garment pieces by end-users. years. Consumers today are more likely to shop Although traditional methods have made important online than to visit a retail store. The situation is progress on these under-constrained problems, much more complicated, however, when it comes to learning-based approaches have shown tremendous buying clothes. People need to know how a garment potential to make a notable impact. Compared to fits on them, how it looks, and how it feels. Digital traditional methods, machine-learning algorithms are try-on systems can potentially satisfy these needs, usually much faster since training and optimization providing a direct visual impression, and possibly are performed offline. They are also good at customized clothes sizing as well. Therefore, it has generalizing to unseen images without the need for drawn much attention as an attractive alternative to tedious data pre-processing. While extensive research improve the user experience and popularize online exists on 2D image learning, machine learning of fashion shopping. highly variable 3D human body shapes is still far However, the technology is still far from practical, from mature, which is the reason why the open issues easy-to-use, and adequate to replace physical try-on. described above remain elusive. Currently, most try-on systems rely on either imageFor each problem listed above, we motivate its editing, copy-pasting, or template demonstrations, importance, provide a problem description, and present state-of-the-art approaches with potential 1 University of Maryland, College Park, MD 20785, USA. for improvements. We believe that solutions to E-mail: J. Liang, [email protected] ( ); M. C. Lin, these challenging problems will lead to significant advances in digital try-on, as well as other areas of [email protected]. Manuscript received: 2020-06-24; accepted: 2020-07-21 e-commerce.
Abstract Digital try-on systems for e-commerce have the potential to change people’s lives and provide notable economic benefits. However, their development is limite
Data Loading...
