Visual Recommendation and Visual Search for Fashion E-Commerce

Recommender systems are historically one of the most successfull and widely known applications of AI, personalized suggestions are nowadays a valuable commercial application of such systems. Many papers have been published in this field, but it is not yet solved; these models still lack state of the art multi-modal capabilities, such as conversational or visual suggestions. In this contribution we present a novel Visual Recommendation module for fashion e-commerces capable of recommending items based on a concept of visual similarity, and a Visual Search module where users can upload a picture of some clothing and search for the most similar ones in a given e-commerce. In conclusion we discuss about the accessibility of these recommender systems for small and medium enterprises, briefly describing our idea of Recommendations-as-a-Service.