TY - GEN
T1 - Higher order matching for consistent multiple target tracking
AU - Arora, Chetan
AU - Globerson, Amir
N1 - Funding Information:
This research was carried out under project number ME 97025 in the framework of the Strategic Research program of The Netherlands Institute for Metals Research in the Netherlands ( www.nimr.nl ). This research was partially financed with a specialization grant of the Flemish Institute for the Promotion of Scientific-Technological Research in Industry (IWT). The authors acknowledge A. Hovestad and A. Bosch of TNO Industrie (Eindhoven, The Netherlands) for the realization of the chromate coatings, H. Winter of Corus (IJmuiden, The Netherlands) for the GDOES measurements, and Andreas Afseth of Alcan International (Banbury Laboratory, UK) for help during the TEM investigation.
PY - 2013
Y1 - 2013
N2 - This paper addresses the data assignment problem in multi frame multi object tracking in video sequences. Traditional methods employing maximum weight bipartite matching offer limited temporal modeling. It has recently been shown [6, 8, 24] that incorporating higher order temporal constraints improves the assignment solution. Finding maximum weight matching with higher order constraints is however NP-hard and the solutions proposed until now have either been greedy [8] or rely on greedy rounding of the solution obtained from spectral techniques [15]. We propose a novel algorithm to find the approximate solution to data assignment problem with higher order temporal constraints using the method of dual decomposition and the MPLP message passing algorithm [21]. We compare the proposed algorithm with an implementation of [8] and [15] and show that proposed technique provides better solution with a bound on approximation factor for each inferred solution.
AB - This paper addresses the data assignment problem in multi frame multi object tracking in video sequences. Traditional methods employing maximum weight bipartite matching offer limited temporal modeling. It has recently been shown [6, 8, 24] that incorporating higher order temporal constraints improves the assignment solution. Finding maximum weight matching with higher order constraints is however NP-hard and the solutions proposed until now have either been greedy [8] or rely on greedy rounding of the solution obtained from spectral techniques [15]. We propose a novel algorithm to find the approximate solution to data assignment problem with higher order temporal constraints using the method of dual decomposition and the MPLP message passing algorithm [21]. We compare the proposed algorithm with an implementation of [8] and [15] and show that proposed technique provides better solution with a bound on approximation factor for each inferred solution.
UR - http://www.scopus.com/inward/record.url?scp=84898826609&partnerID=8YFLogxK
U2 - 10.1109/ICCV.2013.29
DO - 10.1109/ICCV.2013.29
M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???
AN - SCOPUS:84898826609
SN - 9781479928392
T3 - Proceedings of the IEEE International Conference on Computer Vision
SP - 177
EP - 184
BT - Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2013 14th IEEE International Conference on Computer Vision, ICCV 2013
Y2 - 1 December 2013 through 8 December 2013
ER -