Video surveillance cameras with integrated artificial intelligence technology
Intel Movidius has integrated its high-performance, low-power Myriad technology 2 VPU on this manufacturer's safety devices, that can thus use neural networks on the same computer and offer advanced features in analysis and monitoring, among others.
The embedded computer vision technology specialist Intel Movidius has participated with Dahua Technology to publicize the use of its Myriad technology 2 Vision Processing Unit (VPU) in some video surveillance cameras from this manufacturer to offer advanced features in video analysis, such as density monitoring in groups of people; stereoscopic vision; people count, behavioral analysis and detection of illegally parked vehicles.
Movidius technology focuses on delivering high performance based on artificial intelligence (The) with very low-power smart devices. Thanks to Myriad 2 VPU, Dahua can use neural networks on the same video surveillance device, thus bringing intelligence to the cameras, that can be used with or without cloud computing resources.
"The new frontier for AI and machine learning is to use these technologies in real-world products, which means reducing consumer demands to a sufficiently low level to incorporate technologies into the chambers", remi El-Ouazzane said, Vp. of the new technologies group at Intel and CEO of Movidius.
According to Dahua's ad, "Myriad technology 2 VPU provides great computing capacity for deep neural networks, requiring at the same time less than one watt of consumption, allowing us to create some of the smartest products out there to date".
Intel completed the process of acquiring Movidius in November last year, integrating the VPU platform for image processing on the same device as this company, with its depth sensing solution with Intel RealSense technology; a combination that allows autonomous machines to see in three dimensions, understand what surrounds them and act accordingly.
You liked this article?
Subscribe to our RSS feed And you won't miss anything.