International Journal of Scientific Research and Engineering Development

The continuous expansion of online apparel retail has exposed a persistent gap between digital product representation and the customer's ability to evaluate fit, drape, and overall appearance prior to purchase. Conventional storefronts rely on static photographs and tabular size charts, which frequently fail to communicate how a garment will actually behave on an individual body, leading to elevated return volumes and diminished buyer confidence. This work presents an AI-assisted Virtual Trial Room (VTR) that allows shoppers to preview clothing on their own image or live camera feed before checkout. The proposed framework couples a pose-estimation pipeline for skeletal key point extraction with an imageprocessing layer responsible for garment warping, scaling, and overlay alignment, while a lightweight clothbehaviour model approximates fabric drape and stretch under different body postures. A responsive catalogue interface supports category-based browsing, search, and filtering, and an administrative module manages product listings, categories, and user accounts. Functional evaluation across registration, catalogue browsing, image capture, alignment, and preview-saving workflows confirms that the system produces visually coherent overlays with acceptable processing latency on commodity hardware. The resulting platform narrows the gap between in-store and online apparel shopping, offers a practical reduction pathway for size-related returns, and establishes a modular foundation for subsequent enhancements such as threedimensional rendering and AI-driven size recommendation.

Dr.A.Karunamurthy, K. Abishek,"AI-Driven Virtual Trial Room for Online Fashion E-Commerce Using Pose Estimation and Neural Cloth Simulation" International Journal of Scientific Research and Engineering Development, V9(3): Page(2109-2117) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.