Search
Menu
Meadowlark Optics - SEE WHAT

Displays That Look Back

Facebook X LinkedIn Email
Nancy D. Lamontagne, Managing Editor

Digital signage is nothing new. Whether in New York City’s Times Square or along the highway, digital signs enable more than one advertisement to appear in a given location, and the quick changes of the display often catch the attention of passersby. Some digital displays now are capturing not only a viewer’s attention but also data about that viewer.

One company making this possible is a start-up, Quividi of Paris, whose software works with traditional – and inexpensive – webcams to record information about the person viewing an advertisement. The software uses real-time image processing algorithms to extract quantitative and qualitative information from the webcam’s video stream. After preprocessing, each frame is analyzed to detect faces, after which the system tracks how long a person remains within the camera’s field of view.

Display_AEi.jpg

Image courtesy of Quividi.



Paolo Prandoni, founder of Quividi, said, “Our system does not perform a ‘simple counting’ of the people passing in front of the camera – such as what an infrared sensor would do – but can extract more meaningful parameters.” The data can include the number of people who looked – however briefly – toward the camera, the amount of time each person spent in the camera’s field of view (irrespective of the direction in which they are looking), how long the person looked in the direction of the camera and even the viewer’s gender. Prandoni said that the hardest part about developing the technology was getting the sophisticated image processing algorithms to work efficiently on low-end platforms and with low-cost webcams in a variety of lighting conditions. Motomedia of York, UK, has incorporated the technology in advertisements, including one in London for the television show Pushing Daises and one in New York for the television show Andromeda Strain.
Vescent Photonics LLC - Lasers, Combs, Controls 4/15-5/15 MR

Published: October 2008
Consumerdigital signageFeaturesreal-time image processing algorithmsSensors & Detectorswebcams

We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.