Intelligent video surveillance, or predictable future

The scientific information website Phsy.org reported on a major breakthrough made by researchers at Carnegie Mellon University. The results seem to come from the Columbia Broadcasting Corporation’s (CBS) TV series “Person of Interest”. In a study funded by the United States Army Research Laboratory, researchers at Carnegie Mellon University created an artificial intelligence system to observe and predict what a person "may" do in the future. The system uses specially-written software to analyze various real-time video surveillance screens. If unauthorized misconduct is discovered, it will automatically identify and notify the official. According to this report, for example, an airport or station monitor with an automated system can identify a bag that has been abandoned for a few minutes.

This report introduces the complex knowledge of the cognitive engine of the advanced artificial visual intelligence system. It specifically explains how the conceptual parameters of the basic behavior types are determined by the hybrid semantic resources. In a popular sense, this is the behavioral background. For example, does someone drop this bag because he is sitting next to him? Or does the person completely disappear after putting down the bag?

The goal of this research is to create an artificial intelligence system that is similar to human visual intelligence. The computer system is capable of effective and continuous detection. Researchers have noticed that human beings are constantly evolving because they have learned to adapt to the environment and respond appropriately, and are extremely adept at filtering and summarizing sensory data, making decisions, and taking action based on the information and background knowledge acquired. These computer vision algorithms need to rely on more advanced analytical tools, involve the expression and reasoning of knowledge, and are often conducted under uncertain conditions.

The Cognitive Engine is the core component of the Extended Activity Reasoning system in Carnegie Mellon's Mind's Eye program architecture. The “Eye of the Mind” is a project of the US Department of Defense Advanced Research and Technology Agency (DARPA), which aims to create the ability to filter surveillance images, provide support for (far-end) operators, and detect suspicious behavior (such as someone parking Automatically reminds their artificial intelligence system when the field is still running under parcels.

Arizondo, a postdoctoral researcher from the Department of Psychology at Carnegie Mellon University. Alessandro Oltramari and Christian. Christian Lebiere said that this automated video surveillance method can be used for military and civilian purposes.

The Advanced Research Projects Agency (DARPA) is a affiliate of the U.S. Department of Defense and is responsible for research and development of new military technologies.