iccv iccv2013 iccv2013-178 iccv2013-178-reference knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Chen Change Loy, Shaogang Gong, Tao Xiang
Abstract: Regression-based techniques have shown promising results for people counting in crowded scenes. However, most existing techniques require expensive and laborious data annotation for model training. In this study, we propose to address this problem from three perspectives: (1) Instead of exhaustively annotating every single frame, the most informative frames are selected for annotation automatically and actively. (2) Rather than learning from only labelled data, the abundant unlabelled data are exploited. (3) Labelled data from other scenes are employed to further alleviate the burden for data annotation. All three ideas are implemented in a unified active and semi-supervised regression framework with ability to perform transfer learning, by exploiting the underlying geometric structure of crowd patterns via manifold analysis. Extensive experiments validate the effectiveness of our approach.
[1] A. Argyriou, M. Herbster, and M. Pontil. Combining graph laplacians for semi-supervised learning. In NIPS, pages 67–74, 2005. 2
[2] A. Atkinson, A. Donev, and R. Tobias. Optimum experimental designs. Oxford University Press, 2007. 3
[3] M. Belkin and P. Niyogi. Laplacian eigenmaps for dimensionality reduction and data representation. NC, 15(6): 1373–1396, 2003. 5
[4] M. Belkin, P. Niyogi, and V. Sindhwani. Manifold regularization: A geometric framework for learning from labeled and unlabeled examples. JMLR, 7:2399–2434, 2006. 1, 2, 4
[5] C. M. Bishop. Pattern Recognition and Machine Learning. SpringerVerlag, 2007. 3
[6] G. J. Brostow and R. Cipolla. Unsupervised Bayesian detection of independent motion in crowds. In CVPR, pages 594–601, 2006. 2
[7] A. Chan and N. Vasconcelos. Counting people with low-level features and Bayesian regression. TIP, 21(4):2160–2177, 2012. 1, 2, 5, 6, 7
[8] Y. Chang, C. Hu, R. Feris, and M. Turk. Manifold based analysis of facial expression. IVC, 24(6):605–614, 2006. 2
[9] K. Chen, S. Gong, T. Xiang, and C. C. Loy. Cumulative attribute space for age and crowd density estimation. In CVPR, 2013. 1, 2, 5, 7
[14] X. He. Laplacian regularized D-optimal design for active learning and its application to image retrieval. TIP, 19(1):254–263, 2010. 3
[15] D. Helbing, A. Johansson, and E. T. Hochschule. Pedestrian, crowd and evacuation dynamics. Swiss Federal Institute of Technology, 2009. 2
[16] i-LIDS Team. Imagery library for intelligent detection systems (i-
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30] LIDS); a standard for testing video based detection systems. In IEEE International Carnahan Conferences Security Technology, pages 75– 80, 2006. 3, 5 S. Keerthi and D. DeCoste. A modified finite Newton method for fast solution of large scale linear SVMs. JMLR, 6(1):341–361, 2005. 4 K. Kim, F. Steinke, and M. Hein. Semi-supervised regression using Hessian energy with an application to semi-supervised dimensionality reduction. NIPS, 22:979–987, 2009. 2, 4 V. Lempitsky and A. Zisserman. Learning to count objects in images. In NIPS, 2010. 1, 2 M. Li, Z. Zhang, K. Huang, and T. Tan. Estimating the number of people in crowded scenes by mid based foreground segmentation and head-shoulder detection. In ICPR, pages 1–4, 2008. 2 C. C. Loy, K. Chen, S. Gong, and T. Xiang. Crowd counting and profiling: Methodology and evaluation. In Modeling, Simulation, and Visual Analysis of Large Crowds. Springer, 2013. 1, 2, 5 C. C. Loy, T. M. Hospedales, T. Xiang, and S. Gong. Stream-based joint exploration-exploitation active learning. In CVPR, 2012. 3 C. C. Loy, T. Xiang, and S. Gong. Incremental activity modelling in multiple disjoint cameras. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 34(9): 1799–1813, 2012. 2 K. Lu, J. Zhao, and Y. Wu. Hessian optimal design for image retrieval. PR, 44(6):1 155–1 161, 2011. 2, 3 S. Melacci and M. Belkin. Laplacian support vector machines trained in the primal. JMLR, pages 1149–1 184, 2011. 2, 4 A. Y. Ng, M. I. Jordan, and Y. Weiss. On spectral clustering: Analysis and an algorithm. In NIPS, pages 849–856, 2001. 4 T. Ojala, M. Pietikainen, and T. Maenpaa. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. TPAMI, 24(7):971–987, 2002. 5 S. Pan and Q. Yang. A survey on transfer learning. TKDE, 22(10): 1345–1359, 2010. 5 V. Rabaud and S. Belongie. Counting crowded moving objects. In CVPR, pages 705–71 1, 2006. 2 B. Settles. Active Learning. Synthesis Lectures on Artificial Intelli-
[31]
[32]
[33]
[34]
[35] gence and Machine Learning. Morgan & Claypool Publishers, 2012. 3 B. Tan, J. Zhang, and L. Wang. Semi-supervised elastic net for pedestrian counting. PR, 44(10):2297–2304, 2011. 2, 6 S. Vijayanarasimhan, P. Jain, and K. Grauman. Far-sighted active learning on a budget for image and video recognition. In CVPR, pages 3035–3042, 2010. 2 C. Wang. A geometric framework for transfer learning using manifold alignment. PhD thesis, University of Massachusetts Amherst, 2010. 5 C. Wang and S. Mahadevan. Manifold alignment using procrustes analysis. In ICML, pages 1120–1 127, 2008. 2 M.-F. Weng, Y.-Y. Lin, N. C. Tang, and H.-Y. M. Liao. Visual knowledge transfer among multiple cameras for people counting with oc- [1 0]la37HKotec.aDCglheirosnerdna, clirCdzo.ewDdC.SlicnLteo uiayn,rbtmieSnrog.GdeIRolns B.bgMTu,escVtanhCexdn,op2Tme0.r1iXmt2ri.aecn1sgt,a2.4l8,d(3Fe4,s5a)i:g,t5un26r0,feo–7rm5 2i9nu,lt2giv0afr6oi-. [376]rcNYLelI.uPasZiStoeh,aln paishkga-iMenapdsnal1di n6og0Dmr.1ua–IYn1l6detAi-u0PCna8.,MsgkP2-elM0rAoaMn4rca.,io42n0gvS.1exl2Inf.-t3ouUArnmiI,ugplastgpieosnct7rfao3l c–e7ua4str2en,i 2g01.taIsnk8
[132]ITGWmnI.PaCGt ,iVoue1Pno7a,R(bn7Yy,d).:p1mFRaug1.7,ne8C siof–2.l19Did1y8n3lse.–,2arM29na0 in0drkg8,T.2ea0dnH0p9uol.ain2ctga.lprIyomacdegjsu e-btsadfoerdcbhrouwsmtdarencgoaruesn tiosngi.- [398]stTZ2i.0upl Ze8hr. vaiuo2nsm,egRadn.mHsNa.ienvZifachotrila,dw laden adrnBMein.gvWiZrIuon.hamCSnVeg nP.mtRse.,SnptTaePgtciAeorsMan1lI–,mn63d,e02t(hr70ao)cd:8k1s.in92fgo8r–f1s2em1uil,22226633