Researchers Develop Camera That Sees Behind Corners

This post is also available in: עברית (Hebrew)

A company specializing in autonomous vision has announced that its researchers, along with researchers from Princeton University and the University of Montreal, have succeeded to develop a method that allows regular cameras to see objects that are hidden behind walls or corners.

The team of researchers has managed to observe objects hidden out of direct line of sight in high quality and color. The team were able to reconstruct images of 3D objects such as traffic signs.

This was accomplished by using a conventional CMOS camera sensor, the device in modern cameras that convert light into electrons thus creating an image. In order to obtain the image of an object that is not in direct line of sight, the researchers had to use a different illumination method. This is simply done by making a small change to a smartphone’s flash or a car’s headlight.

This new technology has many different applications, mainly in the autonomous vehicle industry. The technology can help self driving cars navigate in situations where the view is limited due to vehicles or any other obstructions.

The technology is also useful for security applications, such as video surveillance.

Benzinga.com mentions that the researcher’s findings will be published at the CVPR computer vision conference, and that they were chosen to present their findings as one of the few oral presenters at the conference.