Microsoft Azure Launches Face Tracking, Emotion Detection In Preview

Azure Media Analytics puts Azure Media Face Detector and emotion detection into preview, can recognise happiness, sadness, anger, fear

Microsoft has released a free public preview of Azure Media Face Detector, a tool that can find and track human faces within a video.

The platform is part of Microsoft’s wider strategy to make its Azure cloud service a comprehensive Platform-as-a-Service product in the hopes of better competing against Infrastructure giants AWS and Google.

‘Multiple faces’

Azure’s ‘Face Detection’ can keep track of multiple faces within in a video (with the most obvious use-case being CCTV), all whilst the faces are moving. During the tracking process, the platform will try to give faces consistent IDs, even the faces are blocked or move off screen temporarily. Time and location metadata is then returned in a JSON file.

It’s important to note that this isn’t face recognition. Microsoft said that an individual who leaves the frame or becomes obstructed for too long will be given a new ID when they return.

CEO Satya Nadella

Microsoft, with the help of machine learning, takes its face detection one step further with emotion detection on its Face Detection Media Processor, too.

Azure’s Emotion Detection returns analysis on multiple ‘emotional attributes’ from the faces detected, including happiness, sadness, fear, and anger.

Last week, Rudiger Dorn, Microsoft’s director of cloud computing, told TechWeekEurope how he sees value in Microsoft’s Platform-as-a-Service (PaaS) offerings such as artificial intelligence and machine learning to track down criminals.

“This is where the value comes, this is what allows you to track criminals and terrorists,” he said, discussing how Azure’s platform competes with public cloud rivals Google and Amazon (AWS).

Azure Media Analytics is Microsoft’s collection of speech and vision services offered at enterprise scale, compliance, security. The services offered as part of Azure Media Analytics are built using the core Azure Media Services platform components.

Writing on a Microsoft blog post announcing the release, Richard Li, program manager for Azure Media Services, said: “This Media Processor (MP) can be used for people counting, movement tracking, and even gauging audience participation and reaction via facial expressions.”

Quiz: Cloud in 2016