2.5D face recognition under tensor manifold metrics

Citation

Chong, Lee Ying and Ong, Thian Song and Chong, Siew Chin (2014) 2.5D face recognition under tensor manifold metrics. In: Neural Information Processing. Lecture Notes in Computer Science . Springer International Publishing, pp. 653-660. ISBN 978-3-319-12643-2

[img] Text
2.5D Face Recognition under Tensor Manifold Metrics.pdf
Restricted to Repository staff only

Download (578kB)

Abstract

Gabor-based region covariance matrix (GRCM) is a very flexible face descriptor where it allows different combination of features to be fused to construct a covariance matrix. GRCM resides on Tensor manifold where the computation of geodesic distance between two points requires the consideration of geometry characteristics of the manifold. Affine Invariant Riemannian Metric (AIRM) is the most widely used geodesic distance metric. However, it is computationally heavy. This paper investigates several geodesic distance metrics on Tensor manifold to find out the alternative speedy method for 2.5D face recognition using GRCM. Besides, we propose a feature-level fusion for 2.5D partial and 2D data to enhance the recognition performance.

Item Type: Book Section
Additional Information: Book Subtitle: 21st International Conference, ICONIP 2014, Kuching, Malaysia, November 3-6, 2014. Proceedings, Part III
Subjects: Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 23 Jan 2015 06:45
Last Modified: 23 Jan 2015 06:45
URII: http://shdl.mmu.edu.my/id/eprint/5945

Downloads

Downloads per month over past year

View ItemEdit (login required)