Long-Term Time-Sensitive Costs for CRF-Based Tracking by Detection

We present a Conditional Random Field (CRF) approach to tracking-by-detection in which we model pairwise factors linking pairs of detections and their hidden labels, as well as higher order potentials defined in terms of label costs. Our method considers long-term connectivity between pairs of detections and models cue similarities as well as dissimilarities between them using time-interval sensitive models. In addition to position, color, and visual motion cues, we investigate in this paper the use of SURF cue as structure representations. We take advantage of the MOTChallenge 2016 to refine our tracking models, evaluate our system, and study the impact of different parameters of our tracking system on performance.

[1]  F. Fleuret,et al.  Multiple object tracking using flow linear programming , 2009, 2009 Twelfth IEEE International Workshop on Performance Evaluation of Tracking and Surveillance.

[2]  Alexandre Heili,et al.  Detection-based multi-human tracking using a CRF model , 2011, 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops).

[3]  François Fleuret,et al.  Exact Acceleration of Linear Object Detectors , 2012, ECCV.

[4]  François Fleuret,et al.  Deformable Part Models with Individual Part Scaling , 2013, BMVC.

[5]  Ramakant Nevatia,et al.  An online learned CRF model for multi-target tracking , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  David A. McAllester,et al.  Object Detection with Discriminatively Trained Part Based Models , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Alexandre Heili,et al.  Exploiting Long-Term Connectivity and Visual Motion in CRF-Based Multi-Person Tracking , 2014, IEEE Transactions on Image Processing.

[8]  Luc Van Gool,et al.  SURF: Speeded Up Robust Features , 2006, ECCV.

[9]  Stefan Roth,et al.  MOT16: A Benchmark for Multi-Object Tracking , 2016, ArXiv.