Head Pose Estimation
Our face detection technology not only locates faces, but it also estimates the three-dimensional head pose (orientation of the head with respect to the camera). Our head pose estimation technology allows end-users to select between coarse and fine-pose estimation. Coarse-pose detection is more computationally efficient, while fine-pose detection is more accurate in orientation. Since pose estimation is an inherent part of face detection, it inherits all its features, including the capability to estimate the pose of low-resolution faces for uncontrolled real-world imagery.
The video below demonstrates the full range of pose estimation for a single individual. As the person on the left moves his head, the 3D head on the right illustrates the corresponding estimated pose.
The video below demonstrates the same capability for a short scene from NBC‘s “The Office.” Note that the size, position and orientation of the detected faces (heads) generates a realistic representation of the relative 3D location of each individual present in the scene.