Generic Face Invariant Model for Face Detection

In this paper we present a model of face class appearance based on learning a relation between features and a face invariant. We have developed a face invariant model for accurate face localization in natural images that presents face pose changes. A probabilistic model capture a relationship between features appearance and invariant geometry is then used to infer a face instance in new image. We use local features which the performances of appearance distinctiveness are sufficient to localize face. An EM classification is applied to determine exactly the face appearance features. Then, invariants parameters are predicted and hierarchical clustering method achieve invariant geometric localization, where clustering deep depends on the aggregate value considered as a factor of precision to construct clusters of invariants. The appearance probabilities of features are computed to select the best cluster and thus to localize face in image. We evaluate our generic invariant by testing it in face detection experiments on PIE, FERET and CMU-Profiles databases. The experimental results show that using face invariant gives a localization rate of 89.3% and results in high accuracy face localization.

[1]  Cordelia Schmid,et al.  Scale & Affine Invariant Interest Point Detectors , 2004, International Journal of Computer Vision.

[2]  Shimon Ullman,et al.  View-Invariant Recognition Using Corresponding Object Fragments , 2004, ECCV.

[3]  Pietro Perona,et al.  Object class recognition by unsupervised scale-invariant learning , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[4]  Luc Van Gool,et al.  SURF: Speeded Up Robust Features , 2006, ECCV.

[5]  David G. Lowe,et al.  Probabilistic Models of Appearance for 3-D Object Recognition , 2000, International Journal of Computer Vision.

[6]  Michael Brady,et al.  Saliency, Scale and Image Description , 2001, International Journal of Computer Vision.

[7]  Dan Roth,et al.  Learning to detect objects in images via a sparse, part-based representation , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Jean-Michel Morel,et al.  A fully affine invariant image comparison method , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[9]  Pietro Perona,et al.  A Bayesian approach to unsupervised one-shot learning of object categories , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[10]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[11]  Matthew Toews,et al.  Detection Over Viewpoint via the Object Class Invariant , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[12]  Antti Oulasvirta,et al.  Computer Vision – ECCV 2006 , 2006, Lecture Notes in Computer Science.

[13]  J.B. Burns,et al.  View Variation of Point-Set and Line-Segment Features , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Cordelia Schmid,et al.  Selection of scale-invariant parts for object class recognition , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[15]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[16]  Jiří Matas,et al.  Computer Vision - ECCV 2004 , 2004, Lecture Notes in Computer Science.