A new camera could prevent companies from collecting embarrassing and identifiable photos and videos from devices like smart home cameras and robot vacuums. It’s called PrivacyLens and was designed by engineers at the University of Michigan.
PrivacyLens uses both a standard video camera and a heat-sensing camera to identify people in images based on their body temperature. The image of the person is then completely replaced by a generic stick figure, whose movements mirror those of the person it represents. The accurately animated stick figure allows a camera-based device to continue to operate without revealing the identity of the person in the camera’s view.
The added anonymity could prevent private moments from being leaked online, which is increasingly common in today’s world of camera-laden devices that collect and upload information. In 2020, a photo of a person using the toilet appeared on an online forum. The person didn’t realize that their iRobot Roomba had wandered into the bathroom and that all of their photos had been sent to a startup’s cloud server. From there, the photos were viewed and shared across social media groups, according to a report investigation by MIT Technology Review.
“Most consumers don’t think about what happens to the data collected by their favorite smart home devices. In most cases, raw audio, images and video files are streamed from these devices to manufacturers’ cloud servers, whether or not the data is actually needed for the end application,” said Alanson Sample, associate professor of computer science and engineering at UM and corresponding author of the study describing the device.
“A smart device that removes personally identifiable information before sensitive data is sent to private servers will be a much more secure product than what we have today.”
Raw photos are never stored on the device or in the cloud, completely eliminating access to unprocessed images. With this level of privacy protection, the engineering team hopes patients will become more comfortable using cameras to monitor their chronic health conditions and fitness at home.
“Cameras provide rich information for monitoring health. They could help track exercise habits and other daily living activities, or call for help when an elderly person falls,” said Yasha Iravantchi, a PhD student in computer science and engineering who will present PrivacyLens on July 18 at 4:30pm BST. Symposium on Privacy Enhancing Technologies in Bristol, United Kingdom
“But this poses an ethical dilemma for people who might benefit from this technology. Without privacy protections, we find ourselves in a situation where they must consider giving up their privacy in exchange for good chronic care. This device could allow us to obtain valuable medical data while preserving patient privacy.”
According to an initial survey of 15 participants, replacing patients with stick figures helps them feel more comfortable with a camera even in the most intimate parts of the home. The team built a sliding privacy scale into the device that lets users control how much of their face and body is censored.
“Our research suggests that people might be comfortable with blurring just their face in the kitchen, but in other rooms of the house, they might want their entire body removed from the image,” Sample said. “We want to give people control over their private information and who has access to it.”
The device could not only make patients more comfortable with chronic monitoring of their health, but it could also help protect privacy in public spaces. Automakers could potentially use PrivacyLens to prevent their autonomous vehicles to be used as surveillance drones, and companies that use cameras to collect data outdoors could find the device useful in complying with privacy laws.
“There are a wide range of tasks where we want to know when people are present and what they are doing, but capturing their identity is not useful for performing the task. So why take the risk?” Iravantchi said.
This research was funded by the Rackham Graduate School and a research gift from the Meta faculty.
Sample has filed a provisional patent for the device, with the help of UM Innovation Partnerships, and hopes to eventually bring it to market.
Study: PrivacyLens: Removing Personally Identifiable Information on Device from RGB Images Using Enhanced Thermal Sensing (DOI: 10.56553/popets-2024-0146)
/Public Release. This material from the original organization/authors may be of a timely nature and edited for clarity, style, and length. Mirage.News takes no institutional position or bias, and all views, positions, and conclusions expressed herein are solely those of the author(s).See full article here.