cvpr cvpr2013 cvpr2013-318 cvpr2013-318-reference knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Sitapa Rujikietgumjorn, Robert T. Collins
Abstract: We present a quadratic unconstrained binary optimization (QUBO) framework for reasoning about multiple object detections with spatial overlaps. The method maximizes an objective function composed of unary detection confidence scores andpairwise overlap constraints to determine which overlapping detections should be suppressed, and which should be kept. The framework is flexible enough to handle the problem of detecting objects as a shape covering of a foreground mask, and to handle the problem of filtering confidence weighted detections produced by a traditional sliding window object detector. In our experiments, we show that our method outperforms two existing state-ofthe-art pedestrian detectors.
[1] N. Dalal and B. Triggs. Histograms of oriented gradients for human detection. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1:886–893, June 2005. 2, 3, 6
[2] C. Desai, D. Ramanan, and C. Fowlkes. Discriminative models for multi-class object layout. IEEE International Conference on Computer Vision (ICCV), pages 229–236, Sept. 2009. 2, 5
[3] P. Dollar, Z. Tu, P. Perona, and S. Belongie. Integral channel features. pages 91.1–91. 11, 2009. 2
[4] P. Dollar, C. Wojek, B. Schiele, and P. Perona. Pedestrian detection: An evaluation of the state of the art. IEEE Transactions on Pattern Analysis and Machine Intelligence, 34(4):743–761, Apr. 2012. 1
[5] I. Dotu, P. Van Hentenryck, M. A. Patricio, A. Berlanga, J. Garca, and J. M. Molina. Real-time tabu search for video tracking association. Principles and practice of constraint programming, pages 21–34, 2009. 2, 5
[6] I. Endres, V. Srikumar, M.-W. Chang, and D. Hoiem. Learning shared body plans. IEEE Computer Vision and Pattern Recognition (CVPR), pages 3130–3137, June 2012. 2
[7] M. Enzweiler and D. Gavrila. Monocular pedestrian detection: Survey and experiments. IEEE Transactions on Pattern Analysis and Machine Intelligence, 3 1(12):2179–2195, Dec. 2009. 2
[8] P. Felzenszwalb and R. Zabih. Dynamic programming and graph algorithms in computer vision. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33(4):721–740, 2011. 2
[9] P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan. Object detection with discriminatively trained part-based models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(9):1627–1645, 2009. 2
[10] W. Ge and R. Collins. Marked point processes for crowd counting. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2913–2920, June 2009. 2, 3, 7
[11] D. Geronimo, A. Lopez, A. Sappa, and T. Graf. Survey of pedestrian detection for advanced driver assistance systems. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(7): 1239–1258, 2010. 2
[12] F. Glover. Tabu search: A tutorial. Interfaces, 20(4):74–94, July 1990. 2, 5
[13] J. Liu, R. Collins, and Y. Liu. Surveillance camera autocalibration based on pedestrian height distributions. British Machine Vision Conference, pages 117. 1–1 17. 11, 2011. 4
[14] Y. Liu, S. Shan, W. Zhang, X. Chen, and W. Gao. Granularity-tunable gradients partition (GGP) descriptors for human detection. IEEE Computer Vision and Pattern Recognition (CVPR), pages 1255–1262, June 2009. 2
[15] N. Meinshausen, P. Bickel, and J. Rice. Efficient blind search: Optimal power ofdetection under computational cost constraints. The Annals of Applied Statistics, 3:38–60, 2009. 5
[16] J. More and D. C. Sorensen. Computing a trust region step. SIAM Journal on Scientific and Statistical Computing, 3:553–572, 1983. 5
[17] G. Palubeckis. Multistart tabu search strategies for the unconstrained binary quadratic optimization problem. Annals of Operations Research, 13 1(1-4):259–282, Oct. 2004. 2, 4, 5
[18] M. Rodriguez, I. Laptev, J. Sivic, and J.-Y. Audibert. Density-aware person detection and tracking in crowds. IEEE International Conference on Computer Vision (ICCV), pages 2423–2430, Nov. 2011. 2, 4, 5
[19] P. Sabzmeydani and G. Mori. Detecting pedestrians by learning shapelet features. IEEE Computer Vision and Pattern Recognition (CVPR), pages 1–8, June 2007. 2
[20] W. R. Schwartz, A. Kembhavi, D. Harwood, and L. S. Davis. Human detection using partial least squares analysis. IEEE International Conference on Computer Vision (ICCV), pages 24–31, Sept. 2009. 1, 2, 6
[21] H. Trinh, Q. Fan, P. Gabbur, and S. Pankanti. Hand tracking by binary quadratic programming and its application to retail activity recognition. IEEE Computer Vision and Pattern Recognition (CVPR), pages 1902–1909, 2012. 2
[22] B. Wu and R. Nevatia. Detection of multiple, partially occluded humans in a single image by bayesian combination of edgelet part detectors. IEEE International Conference on Computer Vision (ICCV), 1:90–97, 2005. 2 333666999644 (a) (b) (c) (d) (e) Figure 8. Comparison of three QUBO variants and two baseline methods. Yellow means correct detection, red is a missed detection, and blue is false positive. (a) Our approach 1: foreground shape covering. (b) Our approach 2: detector confidence score. (c) Our approach 3: hybrid of confidence score and shape covering. (d) OpenCV HOG detector. (e) PLS detector.
[23] iTn. c Zrohwaode adnd sit Ru.at iNoensva.tia IE.EEBa Cyoemsipanute hru Vmisainon se agnmde Pnatat eiorn cRreocwogdneidtio en v i(CroVnPmRe)n,t. 2:4 IE06E–E41 C3o,m 2p0u04te.r 2 Vision and Pat ern Recognition (CVPR), page 459466, June 2003. 2, 3, 7
[24] T. Zhao and R. Nevatia. Tracking multiple humans in 333666999755