METHOD FOR EVALUATING FASHION STYLE USING DEEP LEARNING TECHNOLOGY AND SYSTEM THEREFOR

摘要

Provided are a method for evaluating a fashion style based on deep learning, capable of providing an evaluation result with guaranteed objectivity, and a system thereof. According to the present invention, the method comprises the steps of: inputting one or more images provided from a user terminal; analyzing an object included in the one or more images to extract a plurality of fashion characteristics of a user by using a deep learning algorithm when the user exists in the object; using the deep learning algorithm to extract matching information between the plurality of extracted fashion characteristics; evaluating a fashion style of the user based on the matching information; and transmitting an evaluation result to the user terminal. The fashion characteristics are one of the number of clothes worn by the user, the kind, the color and the style of the clothes, skin tone of the user, and other pieces of information affecting the fashion style. The matching information necessarily includes matching information between the skin tone of the user and one of the fashion characteristics except the skin tone.