Multi-Manifold Modeling For Head Pose Estimation

Xiangyang Liu, Hongtao Lu, Wenbin Li

Abstract:

In this paper, we study the identity-independent head
pose estimation problem, in order to handle the appearance
variations, we consider the pose data lying on multiple man
ifolds. We present a novel manifold clustering method to
construct multiple manifolds each of which characterizes the
underlying subspace of some subjects. We first construct a
set of n-simplexes of subjects by using the similarity of pose
images. Then, we present a supervised method to obtain a
low-dimensional manifold embedding for each n-simplex.
Finally, we propose the K-manifold clustering method, in
tegrating manifold embedding and clustering, to make each
learned manifold with unique geometric structure. The exper
imental results on a standard database demonstrate that our
method is robust to variations of identities and achieves high
pose estimation accuracy.