Chinese students invent ‘invisibility cloak’ that hides humans from security cameras

Ever wondered if Harry Potter's 'invisibility cloak' can actually become a reality, which may allow you to hide not only from the world but also from the security cameras?This may sound unbelievable but some graduate students in China, bridging the gap between science fiction and reality, invented a low-cost and plain-looking coat which is capable of hiding the human body from AI monitored-security cameras during the day as well as at night.

According to a report published by South China Morning Post, the graduate students have developed the InvisDefense coat which is visible to the human eyes but has peculiar patterns all over it which blinds the cameras during the day and releases heat signals at night.

ALSO READ|Greenland was once as green as its name, oldest DNA discovery reveals

Add WION as a Preferred Source

In a creative contest sponsored by Huawei Technologies Co, the product won the first prize.

Science behind the innovation

Professor Wang Zhengof Wuhan University's school of computer scienceoversaw the path-breaking project.

“Nowadays, many surveillance devices can detect human bodies. Cameras on the road have pedestrian detection functions and smart cars can identify pedestrians, roads and obstacles. Our InvisDefense allows the camera to capture you, but it cannot tell if you are human,” said Wang.

In the day, contour recognition and motion recognition is used by the cameras to detect human bodies. The InvisDefense's surface is covered with a specially designed camouflage pattern which interferes with the machine vision's recognition algorithm and blinds the camera effectively so that it fails to identify the human.

ALSO READ|Scientists find ‘unusual, intense blast of energy’ from nearby galaxy

At night, infrared thermal imaging is used by the cameras to track human bodies. The InvisDefense coat's inner surface has irregularly shaped temperature-controlling modules that create an unusual pattern of temperatures and successfully confused the camera.

“The most difficult part is the balance of the camouflage pattern. Traditionally, researchers used bright images to interfere with machine vision and it did work. But it stands out to human eyes, making the user even more conspicuous,” explained Wei Hui, a team member and a PhD student who has worked on the core algorithm.

“We used algorithms to design the least conspicuous patterns that can disable computer vision,” he added. In three months, hundreds of tests were carried out by the team before they finalised the best pattern.

(With inputs from agencies)