JwellAR: An AI-Driven Virtual Jewellery Try-On and Personalized Recommendation System
Research Digest on Engineering Management and Social Innovations · 2026
DOI: 10.46647/6yz5ez88 ↗Kurz gefasst
Fachartikel in der Zeitschrift Research Digest on Engineering Management and Social Innovations, 2026. Verfasst von Ashwini Dhumal und 4 weiteren Autor:innen.
Abstract (Originalsprache)
In the case of online jewellery retail industry, the major challenge in the customer interaction is the lack of possibilities to visualize product on the customer, which results in customer dissatisfaction and high rate of returns. Between these two aspects is the creation and application of an AI-based system that attempts to address this gap, which is referred to as JwellAR. JwellAR uses facial landmark detection in real-time with MediaPipe and Dlib to allow users to have a virtual jewellery try-on. It also incorporates machine learning algorithms, such as Support Vector Machines (SVM) to classify faces based on their shapes and K-Means clustering, to detect skin tones so that it could make highly personalized product recommendations. It is also because the system uses Market Basket Analysis based on Apriori algorithm to propose complementary items, and, thus, the cross-selling opportunities are increased. The architecture is based on lightweight Python FastAPI back-end and an Android front-end, which is free of massive AR engines. The important results are that the accuracy of face shape classification is about 92 percent and the accuracy of the skin tone detection is about 88 percent. The review finds that computer vision, machine learning, and data mining integration can be used to provide a radically different approach to e-commerce, which would greatly enhance user confidence and customization. The next steps involve the introduction of 3D try-ons and the extension of this approach to other related areas such as fashion and cosmetics, which corresponds to the Sustainable Development Goal of trying to innovate the industry.
Zitieren
APA
Dhumal, A., Dawkar, S. S. L., Jadhav, B., Ingole, D., & Yewale, V. (2026). JwellAR: An AI-Driven Virtual Jewellery Try-On and Personalized Recommendation System. Research Digest on Engineering Management and Social Innovations. https://doi.org/10.46647/6yz5ez88
Vancouver
Dhumal A, Dawkar SSL, Jadhav B, Ingole D, Yewale V. JwellAR: An AI-Driven Virtual Jewellery Try-On and Personalized Recommendation System. Research Digest on Engineering Management and Social Innovations. 2026. doi:10.46647/6yz5ez88
BibTeX
@article{Dhumal2026,
author = {Dhumal, Ashwini and Dawkar, Sunayana Sutar Lokesh and Jadhav, Balaji and Ingole, Dhananjay and Yewale, Vinaya},
title = {JwellAR: An AI-Driven Virtual Jewellery Try-On and Personalized Recommendation System},
journal = {Research Digest on Engineering Management and Social Innovations},
year = {2026},
doi = {10.46647/6yz5ez88}
}
EUVIMED