Bridging the Gap Between Online and In-Store Shopping: Fashion Recommendations and Virtual Try-On

The development of extensive clothing databases has significantly advanced clothing recognition and recommendation systems, but existing datasets often suffer from limited annotations and challenges in diverse real-world scenarios. This work leverages the DeepFashion dataset to build an advanced fashion recommendation system with a virtual try-on feature. Similar clothing items are retrieved using ResNet50 for feature extraction and the Nearest Neighbor algorithm, while U2Net is used for image segmentation to support realistic garment visualization on user images. The system is designed to improve online shopping by combining personalized recommendations with a more immersive try-on experience.

In 2024 5th International Conference on Data Intelligence and Cognitive Informatics (ICDICI)