Sciweavers

CRV
2005
IEEE

Head Pose Estimation of Partially Occluded Faces

13 years 9 months ago
Head Pose Estimation of Partially Occluded Faces
This paper describes an algorithm which calculates the approximate head pose of partially occluded faces without training or manual initialization. The presented approach works on low-resolution webcam images. The algorithm is based on the observation that for small depth rotations of a head the rotation angles can be approximated linearly. It uses the CamShift (Continuous adaptive Mean Shift) algorithm to track the users head. With a pyramidal implementation of an iterative Lucas-Kanade optical flow algorithm, a certain feature point in the face is tracked. Pan and tilt of the head are estimated from the shift of the feature point relative to the center of the head. 3D Position and roll are estimated from the CamShift results.
Markus T. Wenzel, Wolfram Schiffmann
Added 24 Jun 2010
Updated 24 Jun 2010
Type Conference
Year 2005
Where CRV
Authors Markus T. Wenzel, Wolfram Schiffmann
Comments (0)