Home BlogDataset Training AI to Analyze Esports

Training AI to Analyze Esports

by Joshua New
by
Esports

Data scientist Farzain Majeed has published an algorithm called DeepLeague, and a training dataset, capable of watching footage of the video game League of Legends and tracking player movements, similar to how sports analytics systems would convert player actions on the field to data. The dataset consists of 100,000 labeled images Majeed generated of League of Legends characters moving around the game map. Majeed made DeepLeague and its data freely available as open source to encourage the development of analytics systems for esports.

Get the data.

1
FacebookTwitterPinterestEmail
Joshua New

Joshua New was a senior policy analyst at the Center for Data Innovation. He has a background in government affairs, policy, and communication. Prior to joining the Center for Data Innovation, Joshua graduated from American University with degrees in C.L.E.G. (Communication, Legal Institutions, Economics, and Government) and Public Communication. His research focuses on methods of promoting innovative and emerging technologies as a means of improving the economy and quality of life.

You may also like

Enabling AI Decision-Making in Autonomous Vehicles

Improving Coal Sorting in China

Enhancing Cybersecurity in Autonomous Cars

Tracking Drought Conditions in Europe

Analyzing Job Automation and Augmentation

Enhancing Peatland Mapping in Alaska

Improving Soccer Game Analytics

Studying Forest Regrowth Patterns

Enhancing Robotic Movements

Studying American Sundown Towns

Show Buttons
Hide Buttons