From Paris to Berlin:
Discovering Fashion Style Influences Around the World
Ziad Al-Halah Kristen Grauman
[Paper]
[Video]
[Data]
[Code]
[Results]
[Media Coverage]
[Related Projects]
We propose the first work to model fashion influence relations among major cities around the world learned from a massive set of social media images.
Abstract
The evolution of clothing styles and their migration across the world is intriguing, yet difficult to describe quantitatively. We propose to discover and quantify fashion influences from everyday images of people wearing clothes. We introduce an approach that detects which cities influence which other cities in terms of propagating their styles. We then leverage the discovered influence patterns to inform a forecasting model that predicts the popularity of any given style at any given city into the future. Demonstrating our idea with GeoStyle---a large-scale dataset of 7.7M images covering 44 major world cities, we present the discovered influence relationships, revealing how cities exert and receive fashion influence for an array of 50 observed visual styles. Furthermore, the proposed forecasting model achieves state-of-the-art results for a challenging style forecasting task, showing the advantage of grounding visual style evolution both spatially and temporally.
Video
Paper
From Paris to Berlin: Discovering Fashion Style Influences Around the World
Ziad Al-Halah and Kristen Grauman
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2020.
[paper]
[supplementary]
[arXiv]
@inproceedings{al-halah2020,
title={From Paris to Berlin: Discovering Fashion Style Influences Around the World},
author={Ziad Al-Halah and Kristen Grauman},
booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
arxivId = {2004.01316},
year={2020}
}
Data
The learned trends for 50 styles in 44 cities used in our work:
Examples of Discovered Fashion Influences
Influence relations among cities
Style influence relations discovered by our model among major European (left) and American (right) cities. The number of chords coming out of a node (i.e. a city) is relative to the influence weight of that city on the receiver. Chords are colored according to the source node color, i.e. the influencer. Click on a city's node to highlight its influence relations.
Influence on global fashion trends
Our model infers the fashion influence exerted by different cities on the global style trends. Asian cities influence on the global trend of several styles is shown below. The width of the connection is relative to the influence weight of that city in relation to other influencer of the same style.
A world map of fashion influence
Fashion influence scores as inferred by our model from everyday images of people from 44 major cities around the world. The size and color of a circle are relevant to the measured average exerted influence of that city on its peers.
Media Coverage
Related Projects
Fashion Forward: Forecasting Visual Style in Fashion
Ziad Al-Halah, Rainer Stiefelhagen and Kristen Grauman
IEEE International Conference on Computer Vision (ICCV), October 2017.
[paper]
[project]
Modeling Fashion Influence from Photos
Ziad Al-Halah and Kristen Grauman
IEEE Transactions on Multimedia (TMM), 2020.
[paper]
[project]