Textile and Apparel Industries

Observing Runway Fashion’s Assortment Migration From 1988 to 2023: Through the Lens of Computer Vision

---

Abstract

of clothing categories usually comes first when planning, with fashion runway images being the essential source of information. To make the image analysis process more robust and time-effective, there is a huge expansion of interest in applying computer vision techniques to solve fashion tasks. While most existing research focuses on a relatively short period of runway images and does not emphasize various clothing categories, this research was positioned to study the clothing category migration of runway images between 1988 and 2023 using convolutional neural networks and transfer learning. A total of 271,335 images were downloaded and categorized into 13 clothing categories. The time series plots indicated the changes in clothing categories over the past 30 years. The peaks and valleys of the plots were further linked to world events to explore the reasons behind them.

Keywords: computer vision, convolutional neural network, runway image, assortment planning, clothing design

How to Cite:

Xia, S., Gu, L. & Ruan, Y., (2024) “Observing Runway Fashion’s Assortment Migration From 1988 to 2023: Through the Lens of Computer Vision”, International Textile and Apparel Association Annual Conference Proceedings 80(1). doi: https://doi.org/10.31274/itaa.17110