I’ve updated the Deface face recognition library that I created last year to take advantage of the new Vector class and some other improvements.
It should be fast enough for some limited real-time use now. For instance, the demo below typically finds my face in < 80ms once it’s aquired it. That’s only about 12fps, but for certain applications that would be sufficient. And I’m sure that the tracking can be improved a lot. There’s no real prediction now as is, nor position averaging, etc.
Allow this to use your webcam if you have one. If you don’t, you’re out of luck.
Based on ideas from OpenCV and the well-known Viola and Jones face detection paper, but NOT associated with either. Deface uses XML encoded classifiers to visually detect and track objects (especially faces) in BitmapData. Compatible with classifiers from OpenCV.
You can see the code in action here:
Source code for the first demo is available on the download page. See DemoCode.