Start Date
9-5-2019 10:30 AM
End Date
9-5-2019 11:45 AM
Document Type
Event
Keywords
Computer Vision, Graphic Design, Logos, AI
Description
Logos are a commonplace element of graphic design employed by most major companies. These companies extensively research the psychological implications of their logos to create a perception of their brand identity. Consumers have been trained to understand or identify relevant information about a company from their logo. A common example is how a consumer can identify a fast food logo by the prevalence of the colors red and yellow. By using a known parameter like the type of industry of a company, this paper explores the possibility of building a computer vision classifier of logos by industry. The source of the training data for this classifier will be the logos of industry leaders, companies who are expected to invest heavily in their brand identity and set the standard many other companies follow. This paper also explores the possibility of using this classifier in a human in the loop scenario to generate new logos or improve existing logos to better match the trends of industry leading graphic design.
DOI
https://doi.org/10.5038/BNFT3825
Using Artificial Intelligence to Understand Industry Trends in Logo Design
Logos are a commonplace element of graphic design employed by most major companies. These companies extensively research the psychological implications of their logos to create a perception of their brand identity. Consumers have been trained to understand or identify relevant information about a company from their logo. A common example is how a consumer can identify a fast food logo by the prevalence of the colors red and yellow. By using a known parameter like the type of industry of a company, this paper explores the possibility of building a computer vision classifier of logos by industry. The source of the training data for this classifier will be the logos of industry leaders, companies who are expected to invest heavily in their brand identity and set the standard many other companies follow. This paper also explores the possibility of using this classifier in a human in the loop scenario to generate new logos or improve existing logos to better match the trends of industry leading graphic design.