关键词: 服装推荐, 判断模型, 优化模型, 计算机视觉, 图像理解

Abstract: Judgement and optimization model for personalized clothing recommendation is proposed, which aims to solve the problem of clothing recommendation algorithm inconsistent with the color characteristics of the user color by combination of four seasons color theory and computer vision. Firstly, the four seasons color judgment model is applied to classify the user color feature based on the input face image. Secondly, the optimization processing model is established, according to the results of the four seasons color judgment model and user’s required style to optimize the processing and get the clothing pre-recommendation results. Finally, the user scoring and feedback mechanism are used to improve the optimization result and get the final recommendation result. Experiments show that this method can realize personalized clothing recommendation which is compatible with user color characteristics with high accuracy in practical application.

Key words: clothing recommendation, judgement model, optimizing model, computer vision, image understanding