USF researchers create algorithm to see behind objects
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Aïda Amer/Axios
Superman's X-ray vision helped him save Metropolis. University of South Florida researchers don't wear capes, but their new technology could similarly help people in the real world.
Why it matters: A new algorithm allows researchers to see around obstacles into hidden areas using a photo from an ordinary camera.
- The technology could eventually be used by law enforcement, search-and-rescue teams, strategic military efforts, and in cars to prevent crashes.
Driving the news: John Murray-Bruce, an assistant professor of computer science and engineering, and his doctoral student, Robinson Czajkowski, in a study published last month wrote their research is the first of its kind to successfully reconstruct a hidden scene in 3D using an ordinary digital camera.
- Previously, ordinary cameras could only be used to create rough 2D reconstructions of small spaces. 3D imaging of hidden scenes required expensive specialized equipment, according to the university.
The intrigue: Murray-Bruce, who had been working on the algorithm since 2017, was motivated to keep developing the technology after getting in a car crash in December.
- "Since then, I have even more fire to make this practical and get it into everyday vehicles, every day life," he told Axios.
How it works: The algorithm pulls information from the photo of faint shadows cast on nearby surfaces to create high-quality reconstruction of the hidden scene.
- Our brains do something similar to see in 3D, reading signals from light that bounces off objects. Surfaces outside our field of vision also contribute light to what we see.
- "The signal's always there, it's what you do with it," Murray-Bruce said. "Our algorithm is trying to find the best signal for what we capture."