Novel ear-assisted 3D face recognition under expression variations
by Paweł Krotewicz
International Journal of Biometrics (IJBM), Vol. 8, No. 1, 2016

Abstract: This paper concerns the novel region-based ear-assisted 3D face recognition based on iterative closest point (ICP) algorithm in face expression changing scenario. The proposed algorithm for 3D face biometric recognition was prepared and tested. As a first contribution, current state-of-the-art in the field of 3D face recognition is presented and the main approaches to the problem are briefly described. Furthermore, all the data processing steps: preprocessing, segmentation, feature extraction and feature comparison are described in detail. As a second contribution, the algorithm behaviour is scrutinised on the DMCSv1 database and the results in the form of DET curves are highlighted in this paper. Also, the comparison with results obtained by means of the algorithm neglecting ear regions is provided. It occurs that ear geometry information added to face as an auxiliary input to iterative closest point can greatly improve recognition results especially in the case of very strong facial expressions. Equal error rate does not exceed 6.25% on arbitrary data subset. In the last section, conclusions are formulated and plans for future work are presented.

Online publication date: Tue, 21-Jun-2016

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Biometrics (IJBM):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com