From Culture to Clothing: Discovering the World Events Behind A Century of Fashion Images
This work addresses the problem of automatically linking cultural events to fashion trends, which is currently a manual and limited process, for fashion historians and trend forecasters.
This paper introduces a data-driven approach to identify cultural factors influencing fashion trends by analyzing large-scale datasets of news articles and vintage photos over a century. The model detects influence relationships between world events and clothing choices, and improves visual style forecasting and photo timestamping on two image datasets.
Fashion is intertwined with external cultural factors, but identifying these links remains a manual process limited to only the most salient phenomena. We propose a data-driven approach to identify specific cultural factors affecting the clothes people wear. Using large-scale datasets of news articles and vintage photos spanning a century, we present a multi-modal statistical model to detect influence relationships between happenings in the world and people's choice of clothing. Furthermore, on two image datasets we apply our model to improve the concrete vision tasks of visual style forecasting and photo timestamping. Our work is a first step towards a computational, scalable, and easily refreshable approach to link culture to clothing.