From Culture to Clothing: Discovering the World Events Behind A Century of Fashion Images

TL;DR

This paper introduces a data-driven, multi-modal model that links cultural events to fashion choices over a century, enhancing understanding of fashion's cultural influences and improving related vision tasks.

Contribution

It presents a novel large-scale, multi-modal statistical approach to automatically identify cultural influences on fashion from news and images, enabling scalable analysis.

Findings

Successfully detected influence relationships between world events and clothing choices

Improved accuracy in visual style forecasting

Enhanced photo timestamping capabilities

Abstract

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.