Modeling of Pose Effects in Oriented Filter Responses for Head Pose Estimation
We propose an approach for view angle invariant recognition of 3D objects, based on modeling the variations of local feature values as function of view angle. In recognition stage we can compute the probabilities for any pixel that there is certain feature in a given pose angle. Any maximum likelihood or posterior based estimation methods can then be applied to infer the objects and their view parameters. We demonstrate the method with piecewise linear model for the pose effects, to recognize the location and pose of a head from the two eyes.
KeywordsElevation Angle View Angle Joint Likelihood Piecewise Linear Model Recognition Stage
- 1.Pentland, A. and Moghaddam, B. and Starner, T.: View-based and Modular Eigenspaces for Face Recognition. Proc. of IEEE Conf. on CVPR, Seattle, WA, 1994.Google Scholar
- 2.Okada, K. and von der Malsburg, C.: Analysis and synthesis of human faces with pose variations by a parametric piecewise linear subspace method. In Proc. of IEEE Conf. on CVPR, Kauai, 2001Google Scholar
- 4.Simoncelli, E. P. and Freeman, W. T.: The Steerable Pyramid: A Flexible Architecture for Multi-Scale Derivative Computation. IEEE Second Int’l Conf on Image Processing. Washington DC, October 1995.Google Scholar
- 5.Kalliomaki, I. and Lampinen, J.: Feature-based inference of human head shapes. In P. Ala-Siuru and S. Kaski, editors, STeP 2002-Intelligence, The Art of Natural and Artificial, Proc. 10th Finnish Artificial Intelligence Conference.Google Scholar
- 6.Wiskott, L., Fellous, J.-M., Krüger, N., von der Malsburg, C.: Face Recognition by Elastic Bunch Graph Matching. In Jain, L.C., Halici, U., Hayashi, I., and Lee, S.B. (eds.), Intelligent Biometric Techniques in Fingerprint and Face Recognition. CRC Press (1999)Google Scholar