University researchers work to develop higher resolution 3D cameras

Thursday, March 31, 2016 - 5:40pm

University of Michigan researchers are developing a higher resolution 3D camera, after receiving a $1.2 million donation from the W. M. Keck Foundation, a charitable organization that supports science, engineering and medical research in the United States.

The camera is under development at the College of Engineering. 3D cameras reconstruct three-dimensional images by taking light in from multiple directions, and could be used in 3D filmmaking and virtual reality.

Though some products already exist in the market, many of them do not have enough resolution to create sharp three-dimensional visuals, researchers noted. Ted Norris, professor of electrical engineering and computer science, said this is because conventional 3D cameras available currently use a microlens array, which results in deteriorated quality of the image.

“The microlens approach involves an inherent trade-off between resolution and the ability to refocus or resolve depth,” Norris said in a press release.

University researchers are aiming to resolve this issue by designing a camera that records light through a series of transparent light detectors created from graphene, the thinnest optical material yet known.

The 3D camera project is inspired by earlier collaborative research between Norris and Zhaohui Zhong, associate professor of electrical engineering and computer science, which led to a graphene photodetectors. Graphene is a another form of carbon, which exists in a lattice one atom thick. University researchers believe this material allows for a greater detection of visible light. 

“We realized that our graphene photodetector can indeed be transparent and at the same time having high sensitivity, something conventional photodetectors cannot offer,” Zhong said.

Building the 3D camera from the graphene layers, however, is not an easy task. Apart from perfecting the graphene photodetectors, researchers must develop the software algorithms necessary to reconstruct the images once they are taken, according to Jeffrey Fessler, professor of electrical engineering and computer science, who is also involved in the project.

“The most challenging aspect of the computational imaging part of this project is the extremely large amount of data needed to represent a 3D light field and to reconstruct it from a stack of 2D measurements collected with the transparent sensors,” he said.

Nevertheless, researchers said the diversity of experiences among the members of the project will help propel the project ahead. Norris is also a professor of applied physics, and Fessler is a professor of applied physics, biomedical engineering and radiology. Fessler specializes in tomography, or creating images through penetrating waves.

“Although the physics of X-ray CT imaging is different from the physics of optical imaging, the mathematical principles of how you reconstruct a 3D scene from 2D images is similar,” Fessler said. “This project will build upon the extensive experience my group has in 3D image reconstruction problems.”

The three faculty members involved in the project said they see great potential in the camera once it is finished. Zhong and Fessler named a few applications, including 3D imaging of live cells with microscopes and patterns of chemiluminescence in the piston cylinders of a combustion engine. The ability to give robots a bionic eye through depth perception is also on their list.

The team is first aiming for a single-lens reflex size camera, but Norris hopes to make it so compact it can fit into a smartphone one day.

“(The realization) depends on the level of financial investment," he said. "Five years is a good guess to have working systems, and 10 years to be widely available."