Lights, camera鈥h never mind (Image: David Malan/Ocean/Corbis)
IT WAS being in London that got . The UK is awash with millions of closed-circuit TV cameras, and as Bahl walked around the city, he realised that much of what the cameras record would never be of interest to anyone.
鈥淚 thought, hey, there鈥檚 so much video streaming that no one鈥檚 looking at,鈥� says Bahl, who is based at Microsoft Research in Redmond, Washington. Resources are being squandered all the time, on installing ubiquitous cameras to record stuff that isn鈥檛 worth keeping. 鈥淲e simply don鈥檛 have the ability to process all those images,鈥� says Paul Schrater at the University of Minnesota in Minneapolis.
Advertisement
Bahl鈥檚 trip provided the inspiration for Vigil, an intelligent camera system to be presented next week at the International Conference on Mobile Computing and Networking in Paris.
Before anyone sees its footage, Vigil looks at every frame, counting the number of objects that might be noteworthy, such as people or car licence plates. Only then does it upload snippets of video, ranked from most to least important, to the cloud.
聯It counts noteworthy objects like people or car licence plates, then ranks video clips accordingly聰
For the last two months, the Vigil team has had the system running at three sites, surveying labs or office hallways in London, Redmond and Madison, Wisconsin. Next year they hope to launch a bigger pilot, perhaps monitoring traffic on a city road.
Bahl imagines that a system like Vigil could one day be used to detect the most exciting spots on the field during football games, switching TV coverage there automatically. It could also tell store managers when customers pick up certain products from their shelves.
Butterfleye, a start-up in San Francisco, is crowdfunding a smart camera to keep tabs on homes when the occupants are away. It starts recording whenever it detects movements or sounds, then notifies the user with a link to the live feed and a note describing what it has seen: 鈥淭oday, Butterfleye saw a person at 8.45 am鈥�, for example.
Smart cameras aren鈥檛 just useful for surveillance. At iniLabs in Zurich, Switzerland, engineers have used IBM鈥檚 TrueNorth chip 鈥� designed to mimic neurons at work in the brain 鈥� to power a vision sensor that behaves like the human retina, responding only to changes in light levels.
Kynan Eng, a co-founder of iniLabs, says the technology is far more efficient because image processing only kicks in when something interesting happens. That makes it useful for long-running research experiments or power-hungry devices.
And another device about to launch, Vidalife, is designed to capture memorable moments around the home. When it senses a human interaction happening before the camera, it makes a recording and saves it for posterity. 鈥淓very camera deserves to be a smart camera,鈥� says Raji Kannan, co-founder of LensBricks, the California-based start-up behind Vidalife.
This article appeared in print under the headline 鈥淲atching with intent鈥�
