Globally Optimal Multi-target Tracking

Many recent advances in multiple target tracking aim at finding a (nearly) optimal set of trajectories within a temporal window. To handle the large space of possible trajectory hypotheses, it is typically reduced to a finite set by some form of data-driven or regular discretization. In this work we propose an alternative formulation of multi-target tracking as minimization of a continuous energy. Contrary to recent approaches, we focus on designing an energy that corresponds to a more complete representation of the problem, rather than one that is amenable to global optimization.

Representative Publications:

Continuous Energy Minimization for Multi-Target Tracking
Anton Milan, Stefan Roth, and Konrad Schindler
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2013
(PDF)

Detection- and Trajectory-Level Exclusion in Multiple Object Tracking
Anton Milan, Konrad Schindler, and Stefan Roth
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Portland, USA, 2013
(PDF)

Contact person: Anton Milan, Konrad Schindler