A Regularized Approach to Feature Selection for Face Detection
In this paper we present a trainable method for selecting features from an overcomplete dictionary of measurements. The starting point is a thresholded version of the Landweber algorithm for providing a sparse solution to a linear system of equations. We consider the problem of face detection and adopt rectangular features as an initial representation for allowing straightforward comparisons with existing techniques. For computational efficiency and memory requirements, instead of implementing the full optimization scheme on tenths of thousands of features, we propose to first solve a number of smaller size optimization problems obtained by randomly sub-sampling the feature vector, and then recombining the selected features. The obtained set is still highly redundant, so we further apply feature selection. The final feature selection system is an efficient two-stages architecture. Experimental results of an optimized version of the method on face images and image sequences indicate that this method is a serious competitor of other feature selection schemes recently popularized in computer vision for dealing with problems of real time object detection.
KeywordsFeature Selection Face Detection Regularize Approach Overcomplete Dictionary Face Detection System
Unable to display preview. Download preview PDF.
- 5.Weston, J., Elisseeff, A., Scholkopf, B., Tipping, M.: The use of zero-norm with linear models and kernel methods. Journal of Machine Learning Research 3 (2003)Google Scholar
- 6.Zhu, J., Rosset, S., Hastie, T., Tibshirani, R.: 1-norm support vector machines. In: Advances in Neural Information Processing SYstems 16, MIT Press, Cambridge (2004)Google Scholar
- 7.Donoho, D.: For most large underdetermined systems of linear equations, the minimal l1-norm near-solution approximates the sparsest near-solution (2004)Google Scholar
- 8.Daubechies, I., Defrise, M., Mol, C.D.: An iterative thresholding algorithm for linear inverse problems with a sparsity constraint. Comm. on Pure Appl. Math. 57 (2004)Google Scholar
- 10.Osuna, E., Freund, R., Girosi, F.: Training support vector machines: an application to face detection. CVPR (1997)Google Scholar
- 11.Schneiderman, H., Kanade, T.: A statistical method for 3D object detection applied to faces and cars. In: International Conference on Computer Vision (2000)Google Scholar
- 12.Mohan, A., Papageorgiou, C., Poggio, T.: Example-based object detection in images by components. IEEE Trans. on PAMI 23(4), 349–361 (2001)Google Scholar
- 13.Ullman, S., Vidal-Naquet, M., Sali, E.: Visual features of intermediate complexity and their use in classification. Nature Neuroscience 5(7) (2002)Google Scholar
- 14.Destrero, A., Mol, C.D., Odone, F., Verri, A.: A regularized approach to feature selection for face detection. Technical Report DISI-TR-07-01, Dipartimento di informatica e scienze dell’informazione, Universita’ di Genova (2007)Google Scholar