The Community for Technology Leaders
RSS Icon
Issue No.07 - July (2013 vol.35)
pp: 1757-1772
W. J. Scheirer , Dept. of Mol. & Cellular Biol., Harvard Univ., Cambridge, MA, USA
A. de Rezende Rocha , Inst. of Comput. (IC), Univ. of Campinas (Unicamp), Campinas, Brazil
A. Sapkota , Dept. of Comput. Sci., Univ. of Colorado, Colorado Springs, CO, USA
T. E. Boult , Dept. of Comput. Sci., Univ. of Colorado, Colorado Springs, CO, USA
To date, almost all experimental evaluations of machine learning-based recognition algorithms in computer vision have taken the form of “closed set” recognition, whereby all testing classes are known at training time. A more realistic scenario for vision applications is “open set” recognition, where incomplete knowledge of the world is present at training time, and unknown classes can be submitted to an algorithm during testing. This paper explores the nature of open set recognition and formalizes its definition as a constrained minimization problem. The open set recognition problem is not well addressed by existing algorithms because it requires strong generalization. As a step toward a solution, we introduce a novel “1-vs-set machine,” which sculpts a decision space from the marginal distances of a 1-class or binary SVM with a linear kernel. This methodology applies to several different applications in computer vision where open set recognition is a challenging problem, including object recognition and face verification. We consider both in this work, with large scale cross-dataset experiments performed over the Caltech 256 and ImageNet sets, as well as face matching experiments performed over the Labeled Faces in the Wild set. The experiments highlight the effectiveness of machines adapted for open set evaluation compared to existing 1-class and binary SVMs for the same tasks.
Training, Testing, Support vector machines, Training data, Face recognition, Face, Object recognition,support vector machines, Open set recognition, 1-vs-set machine, machine learning, object recognition, face verification
W. J. Scheirer, A. de Rezende Rocha, A. Sapkota, T. E. Boult, "Toward Open Set Recognition", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.35, no. 7, pp. 1757-1772, July 2013, doi:10.1109/TPAMI.2012.256
[1] R.P. Duin and E. Pekalska, "Open Issues in Pattern Recognition," Computer Recognition Systems, M. Kurzynski, E. Puchala, M. Wozniak, and A. Zolnierek, eds., pp. 27-42, Springer, 2005.
[2] X. Zhou and T. Huang, "Small Sample Learning during Multimedia Retrieval Using BiasMap," Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2001.
[3] A. Torralba and A.A. Efros, "Unbiased Look at Dataset Bias," Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 1521-1528, 2011.
[4] N. Dalal and B. Triggs, "Histograms of Oriented Gradients for Human Detection," Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2005.
[5] A. Smola, "Learning with Kernels," PhD dissertation, Technische Universität Berlin, Germany, Nov. 1998.
[6] V. Vapnik, Estimation of Dependences Based on Empirical Data. Springer-Verlag, 1982.
[7] B. Schölkopf and A.J. Smola, Learning with Kernels. MIT Press, 2002.
[8] J. Keynes, A Treatise on Probability. Macmillan & Company, Limited, 1921.
[9] N. Shackel, "Bertrand's Paradox and the Principle of Indifference," Philosophy of Science, vol. 74, no. 2, pp. 150-175, 2007.
[10] T. Joachims, "A Support Vector Method for Multivariate Performance Measures," Proc. Int'l Conf. Machine Learning, pp. 377-384, 2005.
[11] P. Phillips, P. Grother, and R. Micheals, "Evaluation Methods on Face Recognition," Handbook of Face Recognition, A. Jain and S. Li, eds., pp. 329-348, Springer, 2005.
[12] L. Fayin and H. Wechsler, "Open Set Face Recognition Using Transduction," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 27, no. 11, pp. 1686-1697, Nov. 2005.
[13] A. Rocha and S. Goldenstein, "From Binary to Multi-Class: Divide to Conquer," Proc. Int'l Conf. Computer Vision Theory and Applications, pp. 323-330, Feb. 2009.
[14] B. Raskutti and A. Kowalczyk, "Extreme Re-Balancing for SVMs: A Case Study," ACM SIGKDD Explorations, vol. 6, no. 1, pp. 60-69, 2004.
[15] C. Park, J.Z. Huang, and Y. Ding, "A Computable Plug-In Estimator of Minimum Volume Sets for Novelty Detection," Operations Research, vol. 58, pp. 1469-1480, Sept. 2010.
[16] D.M. Tax and R.P. Duin, "Support Vector Domain Description," Pattern Recognition Letters, vol. 20, pp. 1191-1199, Nov. 1999.
[17] X. Wang, P. Tino, M.A. Fardal, S. Raychaudhury, and A. Babul, "Fast Parzen Window Density Estimator," Proc. IEEE Int'l Joint Conf. Neural Networks, pp. 3267-3274, 2009.
[18] E. Parzen, "On Estimation of a Probability Density Function and Mode," The Annals of Math. Statistics, vol. 33, pp. 1065-1076, Sept. 1962.
[19] B. Schölkopf, J. Platt, J. Shawe-Taylor, A. Smola, and R. Williamson, "Estimating the Support of a High-Dimensional Distribution," Microsoft Research, Technical Report MSR-TR-99-87, ftp:// tr-99-87.pdf, 1999.
[20] X. Zhou and T. Huang, "Relevance Feedback in Image Retrieval: A Comprehensive Review," Multimedia Systems, vol. 8, no. 6, pp. 536-544, 2003.
[21] Y. Chen, X. Zhou, and T. Huang, "One-Class SVM for Learning in Image Retrieval," Proc. IEEE Conf. Image Processing, pp. 34-37, 2001.
[22] C. Zhang, X. Chen, M. Chen, and S. Ching, "A Multiple Instance Learning Approach for Content Based Image Retrieval Using One-Class Support Vector Machine," Proc. IEEE Int'l Conf. Multimedia and Expo, pp. 1142-1145, 2005.
[23] K. Goh, E. Change, and B. Li, "Using One-Class and Two-Class SVMs for Multiclass Image Annotation," IEEE Trans. Knowledge and Data Eng., vol. 17, no. 10, pp. 1333-1346, Oct. 2010.
[24] J. Hongliang, L. Qingshan, and L. Hanqing, "Face Detection Using One-Class-Based Support Vectors," Proc. IEEE Automated Face and Gesture Recognition, pp. 457-462, 2004.
[25] H. Cevikalp and B. Triggs, "Efficient Object Detection Using Cascades of Nearest Convex Model Classifiers," Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2012.
[26] M. Wu and J. Ye, "A Small Sphere and Large Margin Approach for Novelty Detection Using Training Data with Outliers," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 31, no. 11, pp. 2088-2092, Nov. 2009.
[27] M. Manevitz and M. Yousef, "One-Class SVMs for Document Classification," J. Machine Learning Research, vol. 2, pp. 139-154, Mar. 2002.
[28] C. Forstall, S. Jacobson, and W. Scheirer, "Evidence of Intertextuality: Investigating Paul the Deacon's Angustae Vitae," Literary & Linguistic Computing, vol. 26, no. 3, pp. 285-296, Sept. 2011.
[29] Y. Shen and Y. Yang, "A Novel Data Description Kernel Based on One-Class SVM for Speaker Verification," Proc. Int'l Conf. Acoustics, Speech, and Signal Processing, pp. 489-492, 2007.
[30] H. Kadri, M. Davy, A. Rabaoui, Z. Lachiri, and N. Ellouze, "Robust Audio Speaker Segmentation Using One Class SVMs," Proc. EURASIP European Signal Processing Conf., 2008.
[31] S. Rossignol and O. Pietquin, "Single-Speaker/Multi-Speaker Co-Channel Speech Classification," Proc. Conf. Int'l Speech Comm. Assoc., 2010.
[32] A. Rabaoui, H. Kadri, Z. Lachiri, and N. Ellouze, "One-Class SVMs Challenges in Audio Detection and Classification Applications," EURASIP J. Advances in Signal Processing, 2008.
[33] T. Malisiewicz, A. Gupta, and A.A. Efros, "Ensemble of Exemplar-SVMs for Object Detection and Beyond," Proc. IEEE Int'l Conf. Computer Vision, 2011.
[34] J.R. Khemchandani and S. Chandra, "Twin Support Vector Machines for Pattern Classifications," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 29, no. 5, pp. 905-910, May 2007.
[35] S. Chew, S. Lucey, P. Lucey, S. Sridharan, and J. Cohn, "Improved Facial Expression Recognition via Unit-Hyperplane Classification," Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2012.
[36] P. Bartlett and M. Wegkamp, "Classification with a Reject Option using a Hinge Loss," J. Machine Learning Research, vol. 9, pp. 1823-1840, June 2008.
[37] M. Rohrbach, M. Stark, and B. Schiele, "Evaluating Knowledge Transfer and Zero-Shot Learning in a Large-Scale Setting," Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 1641-1648, 2011.
[38] A. Zweig and D. Weinshall, "Exploiting Object Hierarchy: Combining Models from Different Category Levels," Proc. IEEE Int'l Conf. Computer Vision, pp. 1-8, 2007.
[39] C.H. Lampert, H. Nickisch, and S. Harmeling, "Learning to Detect Unseen Object Classes by Between-Class Attribute Transfer," Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 951-958, 2009.
[40] E. Bart and S. Ullman, "Single-Example Learning of Novel Classes Using Representation by Similarity," Proc. British Machine Vision Conf., pp. 951-958, 2005.
[41] M. Palatucci, D. Pomerleau, G. Hinton, and T.M. Mitchell, "Zero-Shot Learning with Semantic Output Codes," Proc. Neural Information Processing Systems, pp. 1-9, 2009.
[42] L. Wolf, T. Hassner, and Y. Taigman, "The One-Shot Similarity Kernel," Proc. IEEE Int'l Conf. Computer Vision, 2009.
[43] M.G. Quiles, Z. Liang, F. Breve, and A. Rocha, "Label Propagation Through Neuronal Synchrony," Proc. IEEE Int'l Joint Conf. Neural Networks, pp. 2517-2524, July 2010.
[44] G. Heidemann, "Unsupervised Image Categorization," Image and Vision Computing, vol. 23, no. 10, pp. 861-876, Oct. 2004.
[45] C.M. Bishop, Pattern Recognition and Machine Learning, first ed. Springer, 2006.
[46] T. Serre and T. Poggio, "A Neuromorphic Approach to Computer Vision," Comm. ACM, vol. 53, no. 10, pp. 54-61, 2010.
[47] J. DiCarlo and D. Cox, "Untangling Invariant Object Recognition," Trends in Cognitive Science, vol. 11, no. 8, pp. 333-341, Aug. 2007.
[48] T. Onoda, H. Murata, and S. Yamada, "Non-Relevance Feedback Document Retrieval Based on One Class SVM and SVDD," Proc. IEEE Int'l Joint Conf. Neural Networks, pp. 1212-1219, 2006.
[49] P. Lévy and F. Pellegrino, Problemes Concrets d'Analyse Fonctionnelle, vol. 8, Gauthier-Villars Paris, 1951.
[50] C. Chang and C.J. Lin, "LIBSVM: A Library for Support Vector Machines,", 2001.
[51] T. Joachims, "Training Linear SVMs in Linear Time," Proc. 12th ACM SIGKDD Int'l Conf. Knowledge Discovery and Data Mining, pp. 217-226, 2006.
[52] D. Lowe, "Distinctive Image Features from Scale-Invariant Keypoints," Int'l J. Computer Vision, vol. 60, no. 2, pp. 91-110, 2004.
[53] A. Sapkota, B. Parks, W. Scheirer, and T. Boult, "FACE-GRAB: Face Recognition with General Region Assigned to Binary Operator," Proc. IEEE Int'l Workshop Biometrics, 2010.
[54] N. Pinto, J.J. DiCarlo, and D.D. Cox, "How Far Can You Get with a Modern Face Recognition Test Set Using Only Simple Features?" Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2009.
[55] J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, "ImageNet: A Large-Scale Hierarchical Image Database," Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2009.
[56] NIST, NIST/SEMATECH e-Handbook of Statistical Methods, series 33. US GPO, 2008.
32 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool