Outfit Generation and Recommendation -- An Experimental Study

TL;DR

This paper conducts a comprehensive evaluation of various outfit generation and recommendation algorithms, comparing their performance on real-world data, and explores adaptations for personalization, including models like GPT, BERT, and Seq-to-Seq LSTM.

Contribution

It provides the first extensive comparison of multiple algorithms for outfit recommendation, including adaptations for personalization and evaluation of new models like GPT and BERT.

Findings

Personalized models outperform non-personalized ones in user satisfaction.

GPT and BERT show promising results for outfit recommendation tasks.

The study offers insights into model adaptations for better personalization.

Abstract

Over the past years, fashion-related challenges have gained a lot of attention in the research community. Outfit generation and recommendation, i.e., the composition of a set of items of different types (e.g., tops, bottom, shoes, accessories) that go well together, are among the most challenging ones. That is because items have to be both compatible amongst each other and also personalized to match the taste of the customer. Recently there has been a plethora of work targeted at tackling these problems by adopting various techniques and algorithms from the machine learning literature. However, to date, there is no extensive comparison of the performance of the different algorithms for outfit generation and recommendation. In this paper, we close this gap by providing a broad evaluation and comparison of various algorithms, including both personalized and non-personalized approaches, using online, real-world user data from one of Europe's largest fashion stores. We present the adaptations we made to some of those models to make them suitable for personalized outfit generation. Moreover, we provide insights for models that have not yet been evaluated on this task, specifically, GPT, BERT and Seq-to-Seq LSTM.