mikejuk writes: We now have two competing face tracking methods: the new FaceHugger and the award winning Predator that got a lot of attention because it provided Kinect quality object tracking but using just a video camera. You point the camera and draw a box around what you want to track and the box follows the object as it moves around the screen — it is that simple and that powerful. It also turns out to be easy to convert a basic object tracker to a specific face tracker. There is a video to see how good it seems to be. FaceHugger not only sticks to the face it is supposed to be tracking but it refuses to be fooled into tracking the wrong face. You can download an executable to try it out on your own machine but so far the source code hasn’t been released. This is one respect that the Predator algorithm wins out. In tests FaceHugger seems to be as good or better than Predator. Can we resist saying it looks like a "face-off". Link to Original Source
Real Programs don't use shared text. Otherwise, how can they use functions
for scratch space after they are finished calling them?