Techniques for outfit composition for fashion trend analysis

Abstract

Outfit composition is a trending area in fashion industry these days due to the fact that the system can provide intelligent suggestions on which kind of clothing to wear, and analyse trends based on the user's interests. In this paper, we analyse various methods which are used for outfit composition detection and provide a significant review about these methods in order to help researchers to aggregatively study and analyse the trends followed by these methods for prediction of the outfit composition. Standard datasets are also described in this text, which will be helpful in many fields like analysis of outfits, colour-based predictions, and others. We conclude this text with some interesting observations in the field and suggest some further enhancements as well based on our study.