Laser-assisted Camera Images Around Corners

Using reflections from a non-mirrored surface, a new ultrafast time-of-flight imaging technique recovers 3-D shapes hidden from sight, essentially allowing the camera to capture images around corners.

Scientists at MIT’s Media Lab used the ultrafast camera to produce recognizable 3-D images of a wooden figurine and of foam cutouts outside the device’s line of sight. The findings could lead to imaging systems that allow emergency responders to evaluate dangerous environments or to vehicle navigation systems that can negotiate blind turns. The instrument also could be used with endoscopic medical devices to produce images of previously obscure regions of the body.

An image captured around the corner. Using ultrafast illumination and imaging, scientists at MIT analyzed scattered background light and computationally reconstructed an image from it. The camera and light source do not need to see the object direction and can take a picture around a corner. The image shows a projection of a 3-D confidence map of the reconstructed volume. (Image: Velten et al, MIT)

To peer into a room that is outside the system’s line of sight, short bursts of light from a Ti:sapphire laser are fired toward the wall opposite the doorway. The light reflects off the wall and into the room, then bounces around and re-emerges, striking the camera detector, which takes measurements every few picoseconds. Because the light bursts are so short, the system gauges how far the light has traveled by measuring the time it takes to reach the detector.

Light bursts are fired several more times from multiple spots on the wall so that they enter the room at several angles. The detector measures the returning light at various angles to formulate a picture of the room’s geometry by comparing the times at which the returning light strikes different parts of the detector.

Femtosecond lasers formerly were used to produce extremely high speed images of biochemical processes in laboratory settings, where the pulses’ trajectories were carefully controlled.

The experimental setup with the hidden object. (Image: Christopher Barsi and Andreas Velten, MIT Media Lab)

“Four years ago, when I talked to people in ultrafast optics about using femtosecond lasers for room-size scenes, they said it was totally ridiculous,” said Ramesh Raskar, an associate professor at MIT and leader of the new research.

Using hardware in the lab of MIT chemist Moungi G. Bawendi, who collaborated on the project, Andreas Velten, a former postdoctoral associate in Raskar’s group, fired femtosecond bursts of light at an opaque screen. The laser light reflected the light onto objects suspended in front of another opaque panel standing in for the back wall of a room.

The data collected from the ultrafast sensor was processed by algorithms developed by Raskar, Velten, graduate student Otkrist Gupta, Harvard University mathematics postdoc Thomas Willwacher and Ashok Veeraraghavan, an assistant professor of electrical engineering and computer science at Rice University. The team’s image-reconstruction algorithm uses a technique called filter backprojection, which is the basis of CAT scans. Although blurry, the 3-D images produced by the algorithm were easily recognizable.

The math required to knit multiple laser measurements together into a visual image is difficult to achieve, but it builds upon research in related fields, according to Andrew Fitzgibbon, a Microsoft Research principal researcher.

“In computer graphics, you’re making a picture,” Fitzgibbon said. “Applying that math to acquiring a picture is a great idea. There are areas of computer graphics which have used that sort of math.”

In the experiments, Raskar’s group found that problems associated with peering around a corner are similar to using multiple antennas to determine the direction of incoming radio signals. The team hopes to use this insight to improve the image quality that the system produces and to enable it to handle more cluttered visual scenes.

“Coming at it from both ends, from the raw scientific question — because you know, it is kind of a scientific question: ‘Could we see around a corner?’ — to the extreme engineering of it — ‘Can we time these pulses to femtoseconds?’ — that combination, I think, is rare,” Fitzgibbon said.

The work appeared in Nature Communications.

See also: Trillion fps Video: Streak Camera Stops Light

For more information, visit: www.mit.edu