Enhancing E-commerce furniture shopping with AR and AI-driven 3D modeling

Abstract

This research aims to enhance the customer experience in e-commerce furniture portals through the integration of augmented reality (AR), a dynamic pricing tool, and AI-driven 3D modeling. Utilizing machine learning algorithms, specifically SIFT and FLAAN, we develop highly accurate 3D models of furniture. The portal features various filters, including price range and categories, to streamline the shopping experience. The AI-enabled pricing tool, implemented through web scraping techniques on platforms like Amazon using BeautifulSoup, ensures sellers can set competitive and accursate prices, enhancing customer value. The platform allows sellers to request the addition of 3D models for their products. Admins then select the most similar 3D .obj models from the Shapenet dataset and embed AR links into the portal, enabling buyers to visualize furniture in their own space through augmented reality. This capability significantly improves the shopping experience by allowing customers to see how furniture fits in their home environment before purchase. The portal is built using Django for the backend and PostgreSQL for database management, ensuring a fast, secure, and reliable user experience. By combining AR, advanced 3D modeling, AI-driven pricing, and robust backend technologies, this research contributes to the development of a seamless and engaging e-commerce platform for furniture retail.