iccv iccv2013 iccv2013-241 iccv2013-241-reference knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Tianfu Wu, Song-Chun Zhu
Abstract: Many object detectors, such as AdaBoost, SVM and deformable part-based models (DPM), compute additive scoring functions at a large number of windows scanned over image pyramid, thus computational efficiency is an important consideration beside accuracy performance. In this paper, we present a framework of learning cost-sensitive decision policy which is a sequence of two-sided thresholds to execute early rejection or early acceptance based on the accumulative scores at each step. A decision policy is said to be optimal if it minimizes an empirical global risk function that sums over the loss of false negatives (FN) and false positives (FP), and the cost of computation. While the risk function is very complex due to high-order connections among the two-sided thresholds, we find its upper bound can be optimized by dynamic programming (DP) efficiently and thus say the learned policy is near-optimal. Given the loss of FN and FP and the cost in three numbers, our method can produce a policy on-the-fly for Adaboost, SVM and DPM. In experiments, we show that our decision policy outperforms state-of-the-art cascade methods significantly in terms of speed with similar accuracy performance.
[1] Y. Amit, D. Geman, and X. D. Fan. A coarse-to-fine strategy for multiclass shape detection. PAMI, 26(12): 1606–1621, 2004.
[2] G. Blanchard and D. Geman. Hierarchical testing designs for pattern recognition. Ann. Statist. , 33(3): 1155–1202, 2005.
[3] L. D. Bourdev and J. Brandt. Robust object detection via soft cascade. In CVPR, 2005.
[4] S. C. Brubaker, J. Wu, J. Sun, M. D. Mullin, and J. M. Rehg. On the design of cascades of boosted ensembles for face detection. IJCV, 77(1-3):65–86, 2008.
[5] M. Chen, Z. E. Xu, K. Q. Weinberger, O. Chapelle, and D. Kedem. Classifier cascade for minimizing feature evaluation cost. JMLR - Proceedings Track, 22:218–226, 2012.
[6] N. Dalal and B. Triggs. Histograms of oriented gradients for human detection. In CVPR, 2005.
[7] P. Felzenszwalb, R. Girshick, and D. McAllester. Cascade
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19] object detection with deformable part models. In CVPR, 2010. P. Felzenszwalb, R. Girshick, D. McAllester, and D. Ramanan. Object detection with discriminatively trained part based models. PAMI, 32(9): 1627 – 1645, 2010. S. Gangaputra and D. Geman. A design principle for coarseto-fine classification. In CVPR, 2006. T. Gao and D. Koller. Active classification based on value of classifier. In NIPS, 2011. R. Girshick, P. Felzenszwalb, and D. McAllester. Discriminatively trained deformable part models, release 5, 2012. A. Grubb and J. A. D. Bagnell. Speedboost: Anytime prediction with uniform near-optimality. In AISTATS, 2012. I. Kokkinos. Rapid deformable object detection using dualtree branch-and-bound. In NIPS, 2011. C. Lampert, M. Blaschko, and T. Hofmann. Efficient subwindow search: A branch and bound framework for object localization. PAMI, 3 1(12):2129–2142, 2009. H. Masnadi-Shirazi and N. Vasconcelos. Risk minimization, probability elicitation, and cost-sensitive svms. In ICML, 2010. H. Masnadi-Shirazi and N. Vasconcelos. Cost-sensitive boosting. PAMI, 33(2):294–309, 2011. O. Pele and M. Werman. Robust real-time pattern matching using bayesian sequential hypothesis testing. PAMI, 30(8): 1427–1443, 2008. B. P ´oczos, Y. Abbasi-Yadkori, C. Szepesv a´ri, R. Greiner, and N. Sturtevant. Learning when to stop thinking and do something! In ICML, 2009. H. Rowley, S. Baluja, and T. Kanade. Neural network-based
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28] face detection. PAMI, 20(1):23–38, 1998. M. Saberian and N. Vasconcelos. Learning optimal embedded cascades. PAMI, 34(10):2005–2018, 2012. H. Sahbi and D. Geman. A hierarchy of support vector machines for pattern detection. JMLR, 7:2087–2123, 2006. H. Schneiderman. Feature-centric evaluation for efficient cascaded object detection. In CVPR, 2004. J. Sochman and J. Matas. Waldboost - learning for time constrained sequential detection. In CVPR, 2005. X. Song, T. Wu, Y. Jia, and S.-C. Zhu. Discriminatively trained and-or tree models for object detection. In CVPR, 2013. V. Vapnik. Statistical learning theory. Wiley, 1998. P. Viola and M. Jones. Robust real-time face detection. IJCV, 57(2): 137–154, 2004. A. Wald. Sequential Analysis. Wiley, New York, 1947. R. Xiao, H. Zhu, H. Sun, and X. Tang. Dynamic cascades for face detection. In ICCV, 2007. 776600