Jul 21 2017
IntelliVision, a pioneer and leader in AI/Deep Learning video analytics software for Smart Cameras, today announced that the latest version of its face recognition and detection software, which uses a combination of AI, CNN (convolutional neural network) and Deep Learning, has achieved accuracy benchmarks comparable to industry leaders like Google and Facebook.
"Our AI and deep learning video analytics software has always been at the forefront of this technology, as our portfolio of products includes over 100 patents," said Vaidhi Nathan, IntelliVision's CEO. "By sharing the computing load between the camera and the cloud we can maximize the power of Smart Cameras built with all the popular video chipsets."
IntelliVision's "Face Recognizer" product is a highly accurate face recognition and detection solution that detects, recognizes and records people's faces from a camera's field of view. It identifies/verifies one or more people in the scene using a stored database of faces. The product's detection capability allows it to identify and extract human faces from a camera's field of view and record them for future retrieval and forensic analysis. Face Recognizer is a key tool for security professionals at government, commercial and industrial sites.
IntelliVision Face Recognizer scores the following accuracy in the leading public test databases:
LFW: 99.6%
YouTube Faces: 96.5%
MegaFace (with 1000 people/distractors): 95.6%
IntelliVision's Face Recognition product also includes Face Detector and Face Similarity Search capabilities, also using CNN and Deep Learning technologies.
"IntelliVision's portfolio of AI and Deep Learning technology products along with patents is defining the next generation video analytics segment," said Nathan. "Our analytical intelligence evolves as the applications run, picking up more and more nuances and details, and giving us one of the highest accuracy ratings in the industry."