Sign In

Communications of the ACM

ACM TechNews

Computer Vision System Analyzes Human Behavior

View as: Print Mobile App Share:

European researchers, coordinated by the Autonomous University of Barcelona, have developed a cognitive computational system using video cameras and software, which is able to recognize and predict human behavior. Human Expressive Graphic Representation of Motion and their Evaluation in Sequences (HERMES) analyzes human behavior based on three levels of video recording sequences. The information is processed by computer vision and artificial intelligence algorithms, which allow the system to recognize movement trends.

The researchers say that HERMES uses two new ideas in the field of computer vision. First, it can describe the movement captured by the cameras in natural language using simple and precise phrases. Second, the system can analyze and discover potentially unusual behavior and give off warning signals.

Researchers anticipate that HERMES can be used in several different fields, especially intelligent surveillance and accident and crime prevention, as well as marketing and psychology.

From UAB Barcelona
View Full Article


Abstracts Copyright © 2010 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account