Home BlogDataset Teaching Machines to Understand What’s Going On In Videos

Teaching Machines to Understand What’s Going On In Videos

by Joshua New
by
Analyzing video

Researchers at the Massachusetts Institute of Technology’s IBM Watson AI Lab have published a collection of one million labeled video clips called the Moments in Time Dataset to help train AI systems to identify and understand actions in videos. Each video clip is three seconds long and depicts people, animals, objects, and other natural phenomenon in a dynamic scene.

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

Enhancing Energy Efficiency on University Campuses

Advancing Precision Medicine Techniques

Enabling Biometric Recognition

Examining Grassland Grazing Intensity in Mainland China

Understanding Societal Progress

Compiling Agricultural Experiments Data

Training Activity Recognition Models

Identifying Hand Movements

Creating Digital Twins

Testing AI Performance in Multiple Languages

Show Buttons
Hide Buttons