Skip to main content
RISE logo

InFashion - Deep Learning for analysis of fashion related images

Speed and accuracy are vital factors when it comes to analysing trends in the fashion industry. Researchers are now working on a project to speed up the trend insights. AI enables more analytic assessments to be made faster, in order to succeed in a market with increasing competitiveness.

Artificial Intelligence and Deep Learning can make the methods used to conduct trend analyzes in the fashion industry more effective.

Today a big part of trend analysis consists of tracking blogs and social media, but through the project InFashion researchers wants to develop techniques that simplifies and streamlines theses analyses. With the help of semantic segmentation we can automatically identify which garments that are worn by a person in a picture, and exactly which pixels are covered by those garments.

The project combines deep neural networks used for the general domain with techniques suited for fashion images, resulting in data that enables us to train models for other analyzes and to generate synthetic data.

The goal of the project is to explore how to improve image analysis for the fashion industry using deep learning. It will allow fashion companies to be able to work more efficiently and accurately. With better and faster analyzes, you can make smarter investments, make better decisions on production volumes and set fashion trends faster. This will not only lead to economic improvements but also ecological, as better analyzes can lead to less redundant production.



Project name





Västra Götaland Region

RISE role in project

Shared responsibility for AI and Deep Learning

Project start




Stockholms Universitet, The Swedish Fashion Council

Olof Mogren

Contact person

Olof Mogren

Senior Researcher

+46 70 396 96 24

Read more about Olof

Contact Olof
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

* Mandatory By submitting the form, RISE will process your personal data.