International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 13 Issue: 01 | Dec 2026
p-ISSN: 2395-0072
www.irjet.net
AI-Based Virtual Dressing Room System Using Media pipe/Open CV in Python Dr.C.P.Divate1, Ms.N.R.Bhokare2, Shubhankar.S.Kulkarni3 ,Vikram.V.Shirdhone4,Pratik.A.Khot5,Mudaasir.Y.Shaikh6,Balu.N.Waghmode7, Vijay.S.Dodmani8 1Dean, Dept of Computer Engineering, Shri Ambabai Talim Sanstha’s Sanjay Bhokare Group of Institute
Miraj(poly), Maharashtra, India
2Lecturer, Dept of Computer Engineering, Shri Ambabai Talim Sanstha’s Sanjay Bhokare Group of Institute
Miraj(poly), Maharashtra, India
3,4,5,6,7,8Student Dept of Computer Engineering, Shri Ambabai Talim Sanstha’s Sanjay Bhokare Group of Institute
Miraj(poly), Maharashtra, India ---------------------------------------------------------------------***--------------------------------------------------------------------
Abstract - This paper presents a real-time, webcam-based
virtual dressing room that overlays 2D garment assets (PNG with transparency) onto a user’s live video feed using Media Pipe Pose for body alignment and Media Pipe Hands for touch less UI control. Media Pipe Pose provides high-fidelity tracking with 33 body landmarks (with optional segmentation), enabling stable shoulder/torso anchoring and proportional scaling of garments without depth sensors or GPUs . Media Pipe Hand Land marker detects 21 hand landmarks and supports efficient tracking in video/live-stream modes, which makes dwell-based selection and gesture navigation practical at interactive frame rates on commodity hardware. The system targets lower fit uncertainty in online shopping and support hygienic interaction for kiosks and at-home use, aiming to reduce purchase friction and returns caused by fit mismatch (a major driver of e-commerce returns).
Chaudhuryetal. (2019) presented a vision-based human pose estimation system for virtual cloth fitting that uses a standard webcam interface. They computed body joints such as shoulders, elbows, and hips to warp garment images onto the user’s live video, demonstrating that camera-only solutions can support online trial rooms without specialized depth sensors. Their work emphasizes the feasibility of building low-cost virtual trial-rooms for e-commerce portals using classical computer-vision pipelines and pose estimation.
E-commerce return rates remain significant, and fit-related dissatisfaction is a common cause—Shopify cites an average e-commerce return rate of 16.9% in 2024 and notes it can reach up to 30% for some retailers, with fit being a frequent reason for returns. In one survey summary reported by Shopify, 65% of online shoppers said they returned items that didn’t fit, reinforcing the need for better pre-purchase visualization for apparel-like products. A low-cost virtual try-on system built on a standard webcam is especially valuable for budget-conscious students, time-constrained working users, and tier-2/3 city consumers, where physical store trials are inconvenient and contactless experiences are preferred.
An IRJMETS study on “Virtual Try-On System for Fashion E-Retailers” examined how virtual dressing rooms can reduce return rates and increase consumer confidence by allowing customers to visualize clothing before purchase. The authors discussed the integration of AR, computer vision, and recommendation logic to provide more realistic previews and size guidance within fashion platforms. They
Problem statement: Static images and size charts cannot show “how it looks on me” across different body proportions
Impact Factor value: 8.315
Key design goal: “Good-enough realism” and smooth interaction (target ~30 FPS) rather than heavy 3D simulation
Their work highlights that lightweight pose-estimation pipelines can deliver acceptable try-on quality without heavy deep-learning models or expensive GPUs, making virtual try-on more accessible for practical deployment.
1. Introduction
|
Marellietal. (2022) proposed an AI-based virtual try-on web application that overlays garments on users in real time using Media Pipe for pose estimation and web-based rendering. They showed that accurate key point detection enables reasonably precise clothing alignment for static or slow movements, achieving around 85–90% overlay precision on consumer hardware.
Pipe Hands; gesture recognition; OpenCV; alpha blending; e-commerce
© 2026, IRJET
Objective: Provide real-time try-on with simple gesture navigation, running fully on-device for privacy and low latency
2. Literature Review
Keywords - Virtual try-on; Media Pipe Pose; Media
|
ISO 9001:2008 Certified Journal
|
Page 206