The new Azure Media Face Detector lets enterprise users on the cloud track people in videos with real time face detection software.
Microsoft has announced a free public preview of the Azure Media Face Detector and has detailed how the service can be used.
The Azure Media Face Detector is a part of Microsoft cloud based Azure Media Analytics suite, a collection of features focused on vision and speech oriented services.
Targeted to enterprise users, the Face Detector is a Media Processor (MP) that can read facial expressions through audience participation, people counting, and movement tracking.
Users can access Azure Media Face Detector through numerous access points, including APIs, with presets, the free to use Azure Media Services Explorer, or the newly launched Azure portal.
Microsoft has used an official blog post to detail how to use the features, which include Face and Emotion Detection:
Emotion detection is an optional component of the Face Detection Media Processor that returns analysis on multiple emotional attributes from the faces detected, including happiness, sadness, fear, anger, and more. This data is currently returned as an aggregate value of the whole window over a customizable window and interval.
The Face Detection technology can locate and then track human faces and multiple faces in a video, tracking each individual even if they are moving around. Users can also view location metadata in a JSON file, while the software attempts to give consistent IDs to faces even if obstructed in real time.