Using nasal curves matching for expression robust 3D nose recognition

Emambakhsh, M., Evans, A. and Smith, M. (2013) Using nasal curves matching for expression robust 3D nose recognition. In: IEEE Conference on Biometrics: Theory, Applications and Systems (BTAS2013), Washington DC, USA, September 29th - October 2, 2013. Available from: http://eprints.uwe.ac.uk/20812

[img]
Preview
PDF - Submitted Version
2MB

Abstract/Description

The development of 3D face recognition algorithms that are robust to variations in expression has been a challenge for researchers over the past decade. One approach to this problem is to utilize the most stable parts on the face surface. The nasal region’s relatively constant structure over various expressions makes it attractive for robust recognition. In this paper, a new recognition algorithm is introduced that is based on features from the three dimensional shape of nose. After denoising, face cropping and alignment, the nose region is cropped and 16 landmarks robustly detected on its surface. Pairs of landmarks are connected, which results in 75 curves on the nasal surface; these curves form the feature set. The most stable curves over different expressions and occlusions due to glasses are selected using forward sequential feature selection (FSFS). Finally, the selected curves are used for recognition. The Bosphorus dataset is used for feature selection and FRGC v2.0 for recognition. The results show highest recognition ranks than any previously obtained using the nose region: 1) 82.58% rank-one recognition rate using only two training samples with varying expression, for 505 different subjects and 4879 samples; 2) 90.01% and 80.01% when Spring2003 is used for training and Fall2003 and Spring2004 for testing in the FRGC v2.0 dataset, for neutral and varying expressions, respectively.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:nose recognition, nasal curves
Faculty/Department:Faculty of Environment and Technology > Department of Engineering Design and Mathematics
ID Code:20812
Deposited By: Professor M. Smith
Deposited On:31 Jul 2013 08:26
Last Modified:27 Sep 2016 09:58

Request a change to this item

Total Document Downloads in Past 12 Months

Document Downloads

Total Document Downloads

More statistics for this item...