Wardrobe Whiz : AI Powered Wardrobe Management and Shopping Assistant for Young Adults

Abstract

The fashion industry is facing a significant challenge with the rise of fast fashion, leading to wardrobe clutter and environmental concerns. Young adults, in particular, struggle with managing their wardrobes efficiently, often purchasing duplicate items and failing to utilize their clothing to its fullest potential. The lack of a comprehensive tool for wardrobe management exacerbates the problem, resulting in wasteful consumption patterns and a disorganized approach to personal style. To address these challenges, Wardrobe Whiz was developed as a virtual wardrobe assistant leveraging cutting-edge technologies. Utilizing the Segformer model for semantic segmentation, the platform accurately identifies and categorizes clothing items from user-uploaded images, enabling precise inventory management. The Gemini model is employed for feature extraction, capturing intricate fashion details and facilitating sophisticated outfit recommendations based on style compatibility. The recommendation engine, powered by a combination of collaborative filtering and content-based filtering, provides personalized outfit suggestions tailored to occasions and weather conditions, enhancing the user's decision-making process and promoting sustainable fashion choices. Testing and evaluation of Wardrobe Whiz's functionalities have shown promising results, with the Segformer model demonstrating high accuracy in segmenting clothing items. The recommendation engine's effectiveness in providing relevant and diverse outfit suggestions has also been noteworthy. These results underscore the potential of Wardrobe Whiz as a comprehensive solution for modern wardrobe management, combining technological innovation with a commitment to sustainability.