|Video analytics have enabled widespread applications ranging from security surveillance to business intelligence. Although existing surveillance cameras can identify and track people, the cameras' utility is limited by the unavailability of any direct communication between people and the camera.
Researchers at Purdue University have developed a real-time, end-to-end system that can digitally associate people in camera view to their smartphones without knowing their IP/MAC address. This technology enables a camera to send customized messages to a human in its view without prior registration. Such a system has the potential to contribute to ensuring public safety as well as improving the general quality of life.
-Combines context to improve accuracy
-Chooses effective context based on current situation
-Compresses context to only a few bytes
-Public cameras on streets
Mar 2, 2020
Mar 4, 2019
Purdue Office of Technology Commercialization
1801 Newman Road
West Lafayette, IN 47906
Phone: (765) 588-3475
Fax: (765) 463-3486