000164041 001__ 164041
000164041 005__ 20181203035922.0
000164041 0247_ $$2doi$$a10.1109/TPAMI.2011.21
000164041 022__ $$a0162-8828
000164041 02470 $$2ISI$$a000292740000008
000164041 037__ $$aARTICLE
000164041 245__ $$aMultiple Object Tracking using K-Shortest Paths Optimization
000164041 269__ $$a2011
000164041 260__ $$bInstitute of Electrical and Electronics Engineers$$c2011
000164041 336__ $$aJournal Articles
000164041 520__ $$aMulti-object tracking can be achieved by detecting objects in individual frames and then linking detections across frames. Such an approach can be made very robust to the occasional detection failure: If an object is not detected in a frame but is in previous and following ones, a correct trajectory will nevertheless be produced. By contrast, a false-positive detection in a few frames will be ignored. However, when dealing with a multiple target problem, the linking step results in a difficult optimization problem in the space of all possible families of trajectories. This is usually dealt with by sampling or greedy search based on variants of Dynamic Programming which can easily miss the global optimum. In this paper, we show that reformulating that step as a constrained flow optimization results in a convex problem. We take advantage of its particular structure to solve it using the k-shortest paths algorithm, which is very fast. This new approach is far simpler formally and algorithmically than existing techniques and lets us demonstrate excellent performance in two very different contexts.
000164041 6531_ $$aData association
000164041 6531_ $$amultiobject tracking
000164041 6531_ $$aK-shortest paths
000164041 6531_ $$alinear programming
000164041 6531_ $$aMultitarget Tracking
000164041 6531_ $$aAlgorithm
000164041 6531_ $$aPeople
000164041 6531_ $$aScene
000164041 700__ $$aBerclaz, Jerome
000164041 700__ $$aTuretken, Engin
000164041 700__ $$0240254$$aFleuret, Francois$$g146262
000164041 700__ $$0240252$$aFua, Pascal$$g112366
000164041 773__ $$j33$$q1806--1819$$tIEEE Transactions on Pattern Analysis and Machine Intelligence
000164041 8564_ $$s5369000$$uhttps://infoscience.epfl.ch/record/164041/files/top.pdf$$yn/a$$zn/a
000164041 909C0 $$0252087$$pCVLAB$$xU10659
000164041 909C0 $$0252189$$pLIDIAP$$xU10381
000164041 909CO $$ooai:infoscience.tind.io:164041$$pSTI$$pIC$$particle$$qGLOBAL_SET
000164041 917Z8 $$x112366
000164041 937__ $$aEPFL-ARTICLE-164041
000164041 970__ $$aBerclaz_TPAMI_2011/LIDIAP
000164041 973__ $$aEPFL$$rREVIEWED$$sPUBLISHED
000164041 980__ $$aARTICLE