Impact of eye detection error on face recognition performance

The locations of the eyes are the most commonly used features to perform face normalisation (i.e. alignment of facial features), which is an essential preprocessing stage of many face recognition systems. In this study, the authors study the sensitivity of open source implementations of five face recognition algorithms to misalignment caused by eye localisation errors. They investigate the ambiguity in the location of the eyes by comparing the difference between two independent manual eye annotations. They also study the error characteristics of automatic eye detectors present in two commercial face recognition systems. Furthermore, they explore the impact of using different eye detectors for training/enrolment and query phases of a face recognition system. These experiments provide an insight into the influence of eye localisation errors on the performance of face recognition systems and recommend a strategy for the design of training and test sets of a face recognition algorithm.

[1]  Hossein Mobahi,et al.  Toward a Practical Face Recognition System: Robust Alignment and Illumination by Sparse Representation , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Sébastien Marcel,et al.  An Open Source Framework for Standardized Comparisons of Face Recognition Algorithms , 2012, ECCV Workshops.

[3]  Sébastien Marcel,et al.  Bob: a free signal processing and machine learning toolbox for researchers , 2012, ACM Multimedia.

[4]  David J. Kriegman,et al.  Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection , 1996, ECCV.

[5]  Qiang Ji,et al.  Automatic Eye Detection and Its Validation , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[6]  M. Turk,et al.  Eigenfaces for Recognition , 1991, Journal of Cognitive Neuroscience.

[7]  Sébastien Marcel,et al.  Score calibration in face recognition , 2014, IET Biom..

[8]  Takeo Kanade,et al.  Multi-PIE , 2008, 2008 8th IEEE International Conference on Automatic Face & Gesture Recognition.

[9]  Patrick J. Flynn,et al.  Eye Perturbation Approach for Robust Recognition of Inaccurately Aligned Faces , 2005, AVBPA.

[10]  Rolf P. Würtz,et al.  Face Recognition with Disparity Corrected Gabor Phase Differences , 2012, ICANN.

[11]  Klaus J. Kirchberg,et al.  Robust Face Detection Using the Hausdorff Distance , 2001, AVBPA.

[12]  T. Boult,et al.  The eyes have it , 2003, WBMA '03.

[13]  Sébastien Marcel,et al.  Cross-Pollination of Normalization Techniques From Speaker to Face Authentication Using Gaussian Mixture Models , 2012, IEEE Transactions on Information Forensics and Security.

[14]  Avinash C. Kak,et al.  PCA versus LDA , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Sébastien Marcel,et al.  A Scalable Formulation of Probabilistic Linear Discriminant Analysis: Applied to Face Recognition , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Samy Bengio,et al.  Measuring the performance of face localization systems , 2006, Image Vis. Comput..

[17]  Qiang Ji,et al.  Modeling and Predicting Face Recognition System Performance Based on Analysis of Similarity Scores , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Wen Gao,et al.  Local Gabor binary pattern histogram sequence (LGBPHS): a novel non-statistical model for face representation and recognition , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[19]  Joe Marques,et al.  Effects of Eye Position on Eigenface-Based Face Recognition Scoring , 2003 .

[20]  Sébastien Marcel,et al.  Inter-session variability modelling and joint factor analysis for face authentication , 2011, 2011 International Joint Conference on Biometrics (IJCB).

[21]  Sébastien Marcel,et al.  Cross-pollination of normalisation techniques from speaker to face authentication using Gaussian mixture models , 2012 .

[22]  Patrick J. Flynn,et al.  Sensitivity of face recognition performance to eye location accuracy , 2005, SPIE Defense + Commercial Sensing.

[23]  H.K. Ekenel,et al.  Face alignment by minimizing the closest classification distance , 2009, 2009 IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems.

[24]  Matthieu Guillaumin,et al.  Segmentation Propagation in ImageNet , 2012, ECCV.

[25]  Wen Gao,et al.  Curse of mis-alignment in face recognition: problem and a novel mis-alignment learning solution , 2004, Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 2004. Proceedings..