Example based learning for object detection in images

ABSTRACT

In this paper, we describe a general learning architecture for object detection especially car detection. In order to build such a system, we first perform dimension reduction for each example by using maximizing mutual information criterion. The algorithm directly selects projection basis from examples which can minimize Bayes error. This algorithm is named as Maximizing Mutual Information(MMI) method. Given projection basis, all of examples are projected onto these basis and then trained by Support Vector Machine(SVM). This approach can be applied to any object with distinguishable patterns. In test process, we find objects in a image by using our exhaustive search algorithm which is called a Scale based Classifier Activation Map(SCAM). We applied our detection scheme into UIUC car/non-car database [2]. In this experiment we detect 181 cars in 170 images with 200 cars. This result is competitive comparing with other papers [1, 12].

REFERENCES

• 1: Shivani Agarwal , Aatif Awan , Dan Roth, Learning to Detect Objects in Images via a Sparse, Part-Based Representation, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.26 n.11, p.1475-1490, November 2004 [doi>10.1109/TPAMI.2004.108]
• 2: S. Agarwal and D. Roth. Car detection. http://l2r.cs.uiuc.edu/~cogcomp/Data/Car/2002.
• 3: Ella Bingham , Heikki Mannila, Random projection in dimensionality reduction: applications to image and text data, Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining, p.245-250, August 26-29, 2001, San Francisco, California [doi>10.1145/502512.502546]
• 4: S. Canu. Svm and kernel methods matlab toolbox. Perception Systemes et Information, INSA de Rouen, Rouen, France, 2005.
• 5: R. Cendrillon and B. C. Lovell. Real-time face recognition using eigenfaces.
• 6: Navneet Dalal , Bill Triggs, Histograms of Oriented Gradients for Human Detection, Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1, p.886-893, June 20-26, 2005 [doi>10.1109/CVPR.2005.177]
• 7: Fano. Transmission of information:A statistical theory of communications. American Journal of Physics 29:793--794, November 1961.
• 8: Hellman and R. J. M. Probability of error, equivocation, and chernoff bound. IEEE Trans. on Information Theroy 16: 368--372, July 1970.
• 9: Anuj Mohan , Constantine Papageorgiou , Tomaso Poggio, Example-Based Object Detection in Images by Components, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.23 n.4, p.349-361, April 2001 [doi>10.1109/34.917571]
• 10: G. B. N. Goel and A. Ne fian. Face recognition experiments with random projection. In SPIE Defense and Security Symposium (Biometric Technology for Human Identification), Orlando, FL 2005.
• 11: Edgar Osuna , Robert Freund , Federico Girosi, Training Support Vector Machines: an Application to Face Detection, Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97), p.130, June 17-19, 1997
• 12: Jianzhong Fang , Guoping Qiu, Car/Non-Car Classification in an Informative Sample Subspace, Proceedings of the 18th International Conference on Pattern Recognition, p.962-965, August 20-24, 2006 [doi>10.1109/ICPR.2006.356]
• 13: Henry A. Rowley , Shumeet Baluja , Takeo Kanade, Neural Network-Based Face Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.20 n.1, p.23-38, January 1998 [doi>10.1109/34.655647]
• 14: Oliver Sidla , Yuriy Lypetskyy , Norbert Brandle , Stefan Seer, Pedestrian Detection and Tracking for Counting Applications in Crowded Situations, Proceedings of the IEEE International Conference on Video and Signal Based Surveillance, p.70, November 22-24, 2006 [doi>10.1109/AVSS.2006.91]
• 15: Kari Torkkola, Feature extraction by non parametric mutual information maximization, The Journal of Machine Learning Research, 3, 3/1/2003
• 16: UCI. Uci machine learning repository. http://archive.ics.uci. edu/ml/datasets/Statlog+(Landsat+Satellite)1993.
• 17: Qiang Zhu , Mei-Chen Yeh , Kwang-Ting Cheng , Shai Avidan, Fast Human Detection Using a Cascade of Histograms of Oriented Gradients, Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, p.1491-1498, June 17-22, 2006 [doi>10.1109/CVPR.2006.119]

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.5 PATTERN RECOGNITION

General Terms:
Algorithms, Experimentation, Performance

Keywords:
dimensionality reduction, exhaustive search, maximizing mutual information, object detection