Multiple-instance pruning for learning efficient cascade...

Details Multiple-instance pruning for learning efficient cascade... Multiple-instance pruning for learning efficient cascade...

2011-08-30

2011-08-30

Vincent, David R (Department: 2129)

Data processing: artificial intelligence

Knowledge processing system

C706S020000, C706S012000

Reexamination Certificate

active

08010471

ABSTRACT:
A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive
egative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

REFERENCES:
patent: 5793888 (1998-08-01), Delanoy
patent: 6697504 (2004-02-01), Tsai
patent: 6804391 (2004-10-01), Blake et al.
patent: 7024033 (2006-04-01), Li et al.
patent: 7099510 (2006-08-01), Jones et al.
patent: 7139738 (2006-11-01), Philomin et al.
patent: 7164781 (2007-01-01), Kim et al.
patent: 2006/0126938 (2006-06-01), Lee et al.
patent: 2006/0248029 (2006-11-01), Liu et al.
patent: 2007/0011119 (2007-01-01), Thaler
patent: 2007/0019863 (2007-01-01), Ito
Viola et al., P., “Robust Real-time Object Detection”, Second International Workshop on Statistical and Computational Theories of Vision—Modeling, Learning, Computing, and Sampling, Jul. 13, 2001.
Viola et al., P., “Rapid Object Detection using a Boosted Cascade of Simple Features”, IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, 2001.
Viola et al., P., “Fast and Robust Classification using Asymmetric AdaBoost and a Detector Cascade”, Advances in Neural Information Processing Systems, Issue 14, vol. 2, p. 1311-1318, 2002.
Ji et al., P., “Combination of Weak Classifiers”, IEEE Transactions on Neural networks, vol. 8, No. 1, Jan. 1997.
Sochman et al., J.,“Inter-stage Feature Propagation in Cascade Building with AdaBoost”, International Conference on Pattern Recognition, 2004.
Lienhart et al., R., “Empirical Analysis of Detection Cascades of Boosted Classifiers for Rapid Object Detection”, DAGM 2003, LNCS 2781, pp. 297-304, 2003.
Thompson, S., “Pruning boosted classifiers with a real valued genetic algorithm”, Knowledge-Based Systems 12, p. 277-284, 1999.
Shotton et al., J., “TextonBoost: Joint Appearance, Shape and Context Modeling for Multi-class Object Recognition and Segmentation”, ECCV 2006, pp. 1-15, 2006.
Zhang et al., C., “Boosting-Based Multimodal Speaker Detection for Distributed Meetings”, pp. 86-91, IEEE, 2006.
Liu, X., “Generic Face Alignment using Boosted Appearance Model”, IEEE, pp. 1-8, 2007.
Zhang et al., L., “Robust Face Alignment Based on Local Texture Classifiers”, IEEE, pp. 1-4, 2005.
Tu, Z., “Probabilistic Boosting-Tree: Learning Discriminative Models for Classification, Recognition, and Clustering”, IEEE International Conference on Computer Vision, pp. 1-8, 2005.
Jang et al., J., “Evolutionary Pruning for Fast and Robust Face Detection”, 2006 IEEE Congress on Evolutionary Computation, pp. 1-7, Jul. 16-21, 2006.
Bourdev, L., J. Brandt, Robust object detection via soft cascade, IEEE Comp. Society Conf. on Comp. Vision and Pattern Recognition, Jun. 20-25, 2005, pp. 236-243, vol. 2.
Brubaker, S. C., M. D. Mullin, and J. M. Rehg, Towards optimal training of cascaded detectors, Proc. of ECCV, 2006, pp. 325-337.
Friedman, J., T. Hastie, and R. Tibshirani, Additive logistic regression: A statistical view of boosting, Annals of Statistics, 2000, pp. 337-307, vol. 28.
Li, S. Z., L. Zhu, Z. Q. Zhang, A. Blake, H. J. Zhang, and H. Shum, Statistical learning of multi-view face detection, Proc. of the 7th European Conf. on Comp. Vision, 2002.
Luo, H., Optimization design of cascaded classifiers, IEEE Comp. Soc. Conf. on Comp. Vision and Pattern Recognition, CVPR'05, 2005, pp. 480-485, vol. 1.
Nowlan, S. and J. Platt, A convolutional neural network hand tracker, Advances in Neural Information Processing Systems 7, 1995, pp. 901-908, San Mateo, CA.
Rowley, H.A., S. Baluja and T. Kanade, Neural network-based face detection, IEEE Trans. on Pattern Analysis and Mach. Intelligence, Jan. 1998, pp. 23-28, vol. 20, No. 1.
Schapire, R. E., and Y. Singer, Improved boosting algorithms using confidence-rated predictions, Machine Learning, 1999, pp. 297-336, vol. 37.
{hacek over (S)}ochman, J., and J. Matas, Waldboost—Learning for time constrained sequential detection, Proc. of CVPR, 2005.
Sung, K., and T. Poggio, Example-based learning for view-based human face detection, IEEE Trans. on Pattern Analysis and Mach. Intelligence, Jan. 1998, pp. 39-51, vol. 20, No. 1.
Viola, P., and M. J. Jones, Rapid object detection using a boosted cascade of simple features, Proc. of IEEE Conf. on Comp. Vision and Pattern Recognition, 2001.
Viola, P., M. J. Jones, Robust real-time face detection, Int'l J. of Comp. Vision, May 2004, pp. 137-154, vol. 57, No. 2.
Viola, P., J. C. Platt, and C. Zhang, Multiple instance boosting for object detection, Proc. of NIPS, 2006, vol. 18.
Wu, B., H. Ai, C. Huang, and S. Lao, Fast rotation invariant multi-view face detection based on real adaboost, Proc. of IEEE Automatic Face and Gesture Recognition, 2004.
Wu, J., M. D. Mullin, J. M. Rehg, Linear Asymmetric Classifier for cascade detectors, Proceedings of the 22nd Int'l Conf. on Mach. Learning, 2005, vol. 119, pp. 988-995, Bonn, Germany.
Wu, J., J. M. Rehg, and M. D. Mullin, Learning a rare event detection cascade by direct feature selection, Advances in Neural Information Processing Systems, NIPS, 2003.
Xiao, R., L. Zhu, H.-J. Zhang, Boosting chain learning for object detection, 9th IEEE Int'l Conf. on Computer Vision, ICCV'03, 2003, vol. 1.
Michael B. Holmes, U.S. Appl. No. 11/777,471, Notice of Allowance, Apr. 13, 2010.
Wilbert L. Starks, U.S. Appl. No. 11/777,482, Office Action, Mar. 17, 2010.
Yan, R., A. G. Hauptmann, Co-retrieval: A boosted reranking approach for video retrieval, IEEE Proceedings of Vision, Image and Signal Processing, Dec. 9, 2005, pp. 888-895, vol. 152, No. 6, IEEE Computer Society, Washington, DC, USA.
Kauchak, D., J. Smarr and C. Elkan, Sources of success for boosted wrapper induction, J. of Mach. Learning Research, May 2004, pp. 499-527, vol. 5, MIT Press.

Affiliated with

Also associated with

Rating

Multiple-instance pruning for learning efficient cascade... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Multiple-instance pruning for learning efficient cascade..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multiple-instance pruning for learning efficient cascade... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2664368