Authors:
S. Rubin Bose, J. Angelin Jeba, R. Regin, O. Jeba Singh, S. Suman Rajest, Uratchayaphon Nararattananukul

Addresses:
School of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Department of Electronics and Communication Engineering, S.A. Engineering College, Chennai, Tamil Nadu, India. Centre for Academic Research, Alliance University, Bengaluru, Karnataka, India. Department of Research and Development, Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India. Faculty of Business Administration, Ramkhamhaeng University, Bang Kapi, Bangkok, Thailand.

Abstract:

The rapid evolution of e-commerce and digital retail has necessitated innovative solutions to bridge the gap between online and in-store shopping experiences. This paper presents a comprehensive intelligent virtual try-on and fashion recommendation system powered by advanced machine learning techniques, specifically integrating ResNet-50 for feature extraction, HR-Net for high-resolution pose estimation, MediaPipe/OpenPose for real-time pose detection, and Gemini API for conversational fashion assistance. The proposed system addresses critical challenges in online fashion retail, including size uncertainty, style mismatches, and a lack of personalized recommendations. Our methodology combines computer vision, deep learning, and natural language processing to create a seamless virtual try-on experience coupled with intelligent fashion recommendations. Through extensive evaluation on benchmark datasets, including VITON-HD, Fashion- MNIST, and custom datasets, our system demonstrates superior performance in garment fitting accuracy (97.2%), pose estimation precision (95.8%), and recommendation relevance (93.4%) compared to existing solutions. The integration of conversational AI through the Gemini API enhances user engagement by providing contextual styling advice and personalized fashion consultation. This research advances AI-powered fashion technology by presenting a holistic approach that combines multiple state-of-the-art technologies to revolutionize online fashion retail experiences.

Keywords: Fashion Recommendation; Computer Vision; Deep Learning (DL); E-Commerce Market; Fashion Technology; Fashion Consultation; Superior Performance; Holistic Approach.

Received on: 15/06/2025, Revised on: 06/08/2025, Accepted on: 21/10/2025, Published on: 07/03/2026

DOI: 10.69888/FTSFDS.2026.000623

FMDB Transactions on Sustainable Finance and Data Science, 2026 Vol. 1 No. 1, Pages: 39–51

• Views : 74
• Downloads : 22

Download PDF