Virtual Try-On for Jewellery

This repository presents an AI-powered solution to revolutionize the online jewellery shopping experience through virtual try-on capabilities. By integrating computer vision and deep learning techniques, users can try on jewellery items virtually using either uploaded images or their webcam.

Features

• Real-Time Jewellery Detection Utilizes a custom-trained YOLO (You Only Look Once) model to detect and localize jewellery such as earrings, necklaces, and rings in real time.

• Interactive Web Application Offers a user-friendly interface for uploading images or using webcam feeds to simulate jewellery try-ons.

• End-to-End AI Pipeline Covers all stages from model training to deployment, providing a seamless user experience.

Technologies Used

• Python – Core programming language for model development and application logic.
• YOLO – Object detection model used for jewellery localization.
• OpenCV – Library used for real-time image processing and integration.

Setup Instructions

1. Clone the repository:

   git clone https://github.com/pranjaykumar926/VIRTUAL-TRY-ON.git

2. Navigate to the project directory:

   cd VIRTUAL-TRY-ON

3. Install dependencies:

   pip install -r requirements.txt

4. Run the application:

   python app.py

Resources

• YOLO Training Notebook – Details the training process, dataset preparation, and configuration.
• Presentation – Highlights the project scope, technical approach, and results.
• Sample Outputs – Includes example images demonstrating the try-on feature.

Future Work

• Integration with more jewellery types.
• Support for 3D rendering and AR-based previews.
• Enhanced facial/jewellery landmark detection for improved placement accuracy.

---

For additional information and updates, please visit the GitHub Repository.