visage|SDK™ FaceDetect package contains powerful techniques to find faces and facial features in still images in form of a well-documented C++ Software Development Kit. FaceDetect identifies the facial features in still images containing one or more human faces in roughly frontal pose. Most standard image formats such as JPEG, GIF etc. are supported for input images.
The result is a set of MPEG-4 facial features (lips contour, nose, eyes, eyebrows and chin).
The minimal size of the face to be detected is about 50 pixels across. Input images can be color or grayscale (color images are internally converted to grayscale). The algorithm works on near-infrared images. visage|SDK™ contains a sample project with full source code demonstrating the usage of the FaceDetect package to find facial features in images. visage|SDK also includes powerful technologies for head, eyes and facial features tracking in recorded or live video sequences. For full 3D head and facial features tracking see FaceTrack package.