Personalized Attire Recommender System

Abstract

Recommendation systems seek for a ranked list of catalogue images that are comparable to the target catalogue image. These systems are crucial in supporting customers in making purchasing decisions based on product recommendations. Finding a true correlation between similar product recommendations and customer satisfaction will significantly improve a company's sales experience. One of the most difficult tasks in fashion recommendation algorithms is finding related photographs for a given product image. Visual similarity-based personal clothing recommendation system, with data obtained via online scraping from various brands of Pakistani clothes. Visual similarity-based recommendations to build similar products with their features, and we also supplied detailed information about the similar image link, brand link, price, cloth, size, and color. The system is made up of many modules, including a visual search clothing recommender system that employs a pre-trained ResNet-50 on ImageNet to extract features from input images and a Nearest Neighbor backed recommender. Resnet50 extracts information from an input image using a convolutional neural network. The closest neighbor algorithm is used to find the most suitable models based on the input image, and suggestions are presented.