Privacy-Protecting Robotic Cameras Scramble Home Footage

Researchers have created new privacy-preserving robotic cameras to protect user anonymity, as a growing number of smart devices enter the home.

The team, made up of researchers from the Australian Centre for Robotics at the University of Sydney and the Centre for Robotics (QCR) at Queensland University of Technology, developed a new camera design that scrambles visual information before it’s digitized, obscuring details of people and the environments they’re in.

Robots can still use and interpret the still images and videos, though they are unrecognizable to the human eye.

The privacy-preserving cameras have potential applications in smart home devices ranging from baby monitors to robotic vacuum cleaners and even smart fridges. The team said that the technology could be deployed beyond the home to public spaces where privacy is a concern, such as warehouses, hospitals, schools and airports.

"Smart devices are changing the way we work and live our lives, but they shouldn't compromise our privacy and become surveillance tools," said Adam Taras, study co-author. "When we think of 'vision' we think of it like a photograph, whereas many of these devices don't require the same type of visual access to a scene as humans do. 

“They have a very narrow scope in terms of what they need to measure to complete a task, using other visual signals, such as color and pattern recognition.”

While previous efforts to protect visual information stored by a camera have targeted images and videos stored in a computer, the Australian team’s approach alters the image before it’s stored.

"This is the key distinguishing point from prior work which obfuscated the images inside the camera's computer – leaving the images open to attack," said Don Dansereau, study co-author. "We go one level beyond to the electronics themselves, enabling a greater level of protection."

To test its efficacy, the team has released its technology to the research community to try and hack. 

Next, the team said they hope to build physical camera prototypes to test the design in the wild. 

"Current robotic vision technology tends to ignore the legitimate privacy concerns of end-users,” said Niko Suenderhauf, QCR’s deputy director. “This is a short-sighted strategy that slows down or even prevents the adoption of robotics in many applications of societal and economic importance.

“Our new sensor design takes privacy very seriously, and I hope to see it taken up by industry and used in many applications.”