This paper outlines a pioneering research endeavor focused on addressing a prevalent challenge in online clothing shopping: the quest for accurate size and fit. Our innovative approach delves into leveraging artificial intelligence-driven 3D mod-eling techniques to extract precise anthropometric measurements of the human body, utilizing straightforward hardware coupled with sophisticated image processing and computer vision methodologies. The primary objective is to craft a scalable, user-friendly application aimed at streamlining the online clothes shopping experience, eliminating uncertainties for customers and enhancing efficiency. Our proposed methodology amalgamates two distinct anthropometric measurement techniques. Firstly, we employ feature point extraction utilizing the Medi-apipe pose estimation neural network. Additionally, a bespoke tape measurement algorithm harnesses robust image processing tools and libraries. Furthermore, ongoing research delves into employing NERF (Neural Radiance Field) and deep learning techniques to refine the accuracy of our 3D models, specifically targeting intricate measurements such as the Neck and Waist. While extant solutions exist for deriving anthropometric measurements from photographic images, our project pioneers the use of 3D models for precise 1 measurements, potentially revolutionizing the online shopping landscape. This ongoing research endeavors to continually explore novel methodologies, striving to enhance the accuracy and usability of our application.
No takes yet. Share an insight, caveat, or question.
Nasarullah et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: