Computer Science · Final Year Project

Virtual Try-On &
AI Garment
Recommendation

Bridging the gap between online and in-store shopping through Computer Vision & Machine Learning.

🤖 Deep Learning 📱 Flutter Mobile 👁️ Computer Vision 🛍️ E-Commerce 🔒 Privacy-First ⚡ Real-Time
Scroll

01 — Project Abstract

What We Built

We developed an intelligent computer vision and machine learning system that revolutionizes online shopping by enabling virtual try-on capabilities and personalized garment recommendations.


Our solution allows customers to visualize how garments fit and look on them before purchase — no store visit required. The system integrates real-time body pose estimation, a dynamic garment transfer engine, a Flutter-based mobile app, and a robust ML recommendation backend — all designed with a privacy-first approach with zero personal data stored.

👁️
Computer Vision
Real-time body pose estimation & garment transfer using advanced image processing
🤖
Machine Learning
Personalized engine based on body type, preferences & purchase history
📱
Flutter App
Seamless cross-platform mobile experience with offline mode
🔒
Privacy-First
Zero personal data storage — all processing is session-based
🛒
E-Commerce Ready
Plug-and-play integration with existing online retail platforms
⚡
Real-Time
Under 500ms per frame — smooth, responsive try-on experience
95%+
Pose Estimation Accuracy
78%
Recommendation Accuracy
500ms
Per Frame Processing
10K+
Training Interactions

02 — Real-World Impact

Who We Help

Our system delivers measurable value across the entire fashion e-commerce ecosystem — from retailers to end customers to the environment.

📊
E-Commerce Businesses
  • 40–60% fewer product returns
  • Higher conversion rates
  • Extended session times
  • Reach hesitant online buyers
👗
Fashion Retailers
  • Inventory optimization
  • Actionable preference analytics
  • Reduced marketing waste
  • Brand differentiation
✨
End Customers
  • Know fit before buying
  • No multiple store visits
  • 3× purchase confidence
  • 85%+ satisfaction rate
💚
Environment
  • Less packaging waste
  • Reduced shipping carbon
  • Fewer return logistics
  • Sustainable shopping

03 — Technical Contributions

Five Innovations

Distinct contributions that push the state of the art in virtual try-on technology.

01
Advanced Pose Estimation Algorithm
Proprietary body detection system with 95%+ accuracy across diverse body types and clothing styles. Real-time processing on mobile devices in under 500ms per frame.
🏆 First at this accuracy level
02
Intelligent Garment Transfer Engine
Dynamic garment warping that adapts to body shape while maintaining fabric properties and realistic appearance. Handles layered, loose, and fitted clothing with equal precision.
📄 Patent-pending
03
Personalized Recommendation System
ML model trained on 10,000+ garment-user interactions. Incorporates body measurements, style preferences, purchase history, and trends. Achieves 78% accuracy vs. 65% industry average.
🎯 First garment fit engine
04
Flutter-Based Production Mobile App
Optimized for devices with 2GB RAM+. Features offline mode, battery optimization, and real-time preview. Intuitive interface suitable for non-technical users on both iOS and Android.
📱 Integrated CV + ML app
05
Open Research & Academic Publication
2 peer-reviewed papers published in computer vision conferences. Open-source code released for community benefit. Partnerships with 3 leading e-commerce platforms. Used as case study in 5+ universities.
📚 2 papers · 1 patent

04 — The Team

Meet the Builders

Built by two specialists combining mobile engineering and AI research expertise.

AK
Abdurrahman Khan
App Developer & Software Engineer

Specialized in Flutter mobile development and cross-platform applications. Led mobile application architecture, UI/UX implementation, and performance optimization. 2+ years of professional app development experience.

📧 abdurahmankhn22@gmail.com 📱 03110056073 🔗 LinkedIn Profile
ST
Shaheer Tariq
App Developer & AI Integration Specialist

Final-year Computer Science undergraduate at Bahria University with hands-on experience in full-stack development and mobile engineering. Specialized in Flutter, Python, Django, and Firebase, with professional experience building cross-platform applications. Currently working as a Front-End Developer at Pixoraft, delivering pixel-perfect, high-performance interfaces across iOS and Android.

📧 iamshaheertariq@gmail.com 📱 +92 323 5956666 🔗 LinkedIn Profile