Facial detection is a big emerging technology in the smartphone space and Microsoft Research Asia is working on staying at the cutting edge. Microsoft scientists released a software development kit this week that includes several of Microsoft’s latest face detection algorithms, including a face detection system that “tries to detect every face that appears in an image” and “identifies the face position of individuals pictured in the image” — a face alignment feature that can locate the individual elements of each face, including the eyes, mouth and nose; and, most intriguingly, a face tracking feature that can find individual faces during live video streams and thus let users “use head movement to interact with a Windows Phone.”
Microsoft has released three Windows Phone apps based on the algorithms so far: Face Swap, which lets users swap faces in and out of different bodies; Face Mask, which lets users cover up faces in pictures for people concerned about their privacy; and Face Touch, which lets users alter facial expressions in pictures just through touching them on the screen.
Qiufeng Yin, a software-development engineer at Microsoft Research Asia in Beijing, said this week that the SDK only supports general facial recognition — that is, recognizing an object as a face with eyes, nose, mouth, etc. — and cannot be used to identify individual faces.
“We are sure people can find many noble and legal uses for face-recognition technology—and have ways to mitigate the side effects—if such technology becomes widely available,” he said. “Over the long term we hope to see a wide spectrum of different applications taking advantage of the unique advantage of human face interaction, especially in the area of digital entertainment.”