Portfolio

OUR PROJECTS

FASHION TAGGER

Supported by TÜBİTAK, this project aims to automate product tagging for the fashion industry using advanced deep learning and image processing techniques. The project involves segmenting images to identify product boundaries and classifying them into main categories. The system determines the dominant color and further classifies products based on specific attributes like skirt length or collar type, generating detailed and grammatically coherent product labels. These labels can be produced in multiple languages, including Turkish and English. This comprehensive tagging system not only reduces the workload for e-commerce sites but also enhances product tracking and control while optimizing labels for search engines.

TWIN SHOP

Supported by KOSGEB, TwinShop revolutionizes fashion discovery by allowing users to find visually similar items with ease. Simply upload an image, and our advanced AI-powered engine will search through our extensive database to find matching or similar fashion products. This seamless experience saves time and helps users explore new styles from various brands. Enhance your shopping journey with accurate and relevant matches based on visual attributes, ensuring you stay trendy and stylish. TwinShop not only improves the shopping experience but also brings innovation to the fashion industry by leveraging cutting-edge technology.