DEEP LEARNING BASED CLOTHING MATCHING: A SURVEY

Abstract

In modern society, clothing has gradually become an important beauty-enhancing product, playing an important role in human’s social life. Nevertheless, each outfit is composed of items of many different types (e.g. top, bottom, shoes) that share some complicated relationship with one another, how to make a suitable outfit has become a daily headache for many people, especially those who do not have much sense of aesthetics. In recent years, the booming fashion sector and its huge potential benefits have attracted tremendous attention from many research communities. Moreover, deep learning has garnered considerable recognition by researchers in many research fields such as computer vision and natural language processing, owing not only to stellar performance but also the attractive property of learning feature representations from scratch. The influence of deep learning is also pervasive, recently demonstrating its effectiveness when applied to information retrieval and outfit recommender systems research. In the light of this, many research efforts have been dedicated to the task of complementary clothing matching and have achieved great success relying on the advanced data-driven neural networks. However, finding suitable architectures, the selection and extraction of clothing features is a challenge for researchers. This article aims to provide a comprehensive review of recent research efforts on deep learning based clothing matching. More concretely, we first review the research on the clothing matching system and categorized deep learning-based clothing matching methods into two categories, namely collaborative method and content-based method. We then carry out a detailed introduction to this relevant research on deep learning based clothing matching.