Researchers at the Stanford University are attempting to develop a system which can produce images of objects that are hidden from view. The researchers are focusing on applications for autonomous vehicles which have laser-based systems in order to detect obstacles or objects around the car which are hidden from view. Other uses of this sensitive technology is seeing through foliage from aeroplanes or aerial vehicles, giving rescue teams the ability to find people blocked from the view by rubble or walls. The idea of non-line-of-sight imaging sounds like magic is actually feasible. A substantial challenge in non-line-of-sight imaging is finding out the best efficient way to recover the 3D structure of the hidden object from noisy environment.
In the system, a laser is set next to a sensitive photon detector which can record even a single particle of light. Pulses of laser light are shot at the wall and these boxes of light which are invisible to the human eye of the objects around the corner and comes back to the wall and to the detector. The scanning time is between two minutes to an hour depending on conditions such as reflective of the hidden object as well as the lighting of the environment. The researchers feel that the computation algorithm is already ready for the LIDAR systems but the real question is whether the LIDAR systems with current hardware support this type of imaging.