acm-header
Sign In

Communications of the ACM

ACM TechNews

Stanford Camera Can Watch Moving Objects Around Corners


View as: Print Mobile App Share:
Part of a demonstration of the non-line-of-sight camera.

A camera system developed by researchers at Stanford University can reassemble scenes and detect moving objects hidden around corners.

Credit: Stanford Computational Imaging Lab

Stanford University researchers have developed a camera system that can reassemble scenes and detect moving objects hidden around corners by analyzing the reflections of individual light particles.

The researchers hope such solutions could help boost the safety of autonomous vehicles and robots.

The camera's hardware, scanning/image-processing speeds, and imaging style already can be found in many self-driving car vision systems.

A powerful laser scans a wall opposite the scene of interest at four frames a second, causing light to bounce off the wall, strike objects in the scene, and bounce back as specks to the camera sensor. The sensor captures each speck and transmits it to an algorithm, which untangles them to reconstruct the concealed scene.

Said Stanford’s David Lindell, “With this hardware, we can basically slow down time and reveal these tracks of light. It almost looks like magic.”

From Stanford News
View Full Article

 

Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account