Maintaining some semblance of privacy in a society where cameras are increasingly prevalent is becoming harder and harder. But the unwanted eyes aren’t just a problem out in public – nowadays the monitoring is coming from inside the house, too. Be it obvious sources like a baby monitor or computer webcam, or more easily overlooked cameras like those in an autonomous vacuum cleaner, we're spending more and more time being watched by our technology.
Which is fine, until your Roomba sees you on the toilet, and the images somehow end up on Facebook.
Now, engineers at the University of Michigan (U-M) have found a way to preserve privacy without sacrificing security or the capabilities of a multitude of smart devices designed to make our lives easier. They’ve developed a new camera dubbed PrivacyLens that can mask a person’s identity by turning them into a stick figure, before it gets sent anywhere.
The PrivacyLens consists of a standard video camera and a heat-sensing camera – to identify the presence of a person from their body temperature. Onboard electronics then remove the person from the video and replace them with an animated stick figure that mimics their actual movements. The processing to achieve this takes place on the camera, so in theory, no identifiable images are transmitted to the cloud.
![Yasha Iravantchi looks like an anonymous stick figure in this monitor connected to PrivacyLens](https://assets.newatlas.com/dims4/default/c820cbb/2147483647/strip/true/crop/1707x2560+0+0/resize/640x960!/quality/90/?url=http%3A%2F%2Fnewatlas-brightspot.s3.amazonaws.com%2Fb7%2F45%2F599738d345d189470894c70708b4%2Fprivacylens2.jpg)
As well as preventing embarrassing breaches of privacy, the Michigan team says its technology would allow users with chronic conditions to take advantage of smart devices that monitor health or exercise routines without fear.
“Cameras provide rich information to monitor health,” said Yasha Iravantchi, a doctoral student in computer science and engineering at U-M. “It could help track exercise habits and other activities of daily living, or call for help when an elderly person falls. But this presents an ethical dilemma for people who would benefit from this technology. Without privacy mitigations, we present a situation where they must weigh giving up their privacy in exchange for good chronic care. This device could allow us to get valuable medical data while preserving patient privacy.”
![Alanson Sample, an associate professor of computer science and engineering, and Yasha Iravantchi, a doctoral student in computer science and engineering, prepare PrivacyLens for a demonstration](https://assets.newatlas.com/dims4/default/9812fe6/2147483647/strip/true/crop/2000x1334+0+0/resize/1439x960!/quality/90/?url=http%3A%2F%2Fnewatlas-brightspot.s3.amazonaws.com%2F6b%2F2e%2Ff803308d41d597c491db06348aad%2Fprivacylens3.jpg)
The researchers have also built a sliding privacy scale into the device that allows users to control the level of censorship. This was in response to a survey that found the desired level of censorship could vary depending on the specific location in the home.
“Our survey suggested that people might feel comfortable only blurring their face when in the kitchen, but in other parts of the home they may want their whole body removed from the image,” said Alanson Sample, U-M associate professor of computer science and engineering and the corresponding author of the study describing the device. “We want to give people control over their private information and who has access to it.”
In addition to adding a welcome level of privacy to devices in the home, the researchers believe their technology could also find use in public spaces where capturing the identity of people isn’t required, but it's still important to know where people are and what they’re doing – be it for cameras on autonomous vehicles or companies using cameras to collect data outdoors.
Sample has filed a provisional patent for PrivacyLens with the goal of a commercial release, while Iravantchi will present the device at the Privacy Enhancing Technologies Symposium in Bristol, U.K. this week.
Source: University of Michigan