Detecting tracking errors via forecasting

Obaidullah Khalid, Andrea Cavallaro and Bernhard Rinner

Abstract

We propose a tracker-independent framework to determine time instants when a video tracker fails. The framework is divided into two steps. First, we determine tracking quality by comparing the distributions of the tracker state and a region around the state. We generate the distributions using Distribution Fields and compute a tracking quality score by comparing the distributions using the L1 distance. Then, we model this score as a time series and employ the Auto Regressive Moving Average method to forecast future values of the quality score. A difference between the original and forecast returns an error signal that we use to detect a tracker failure. We validate the proposed approach over different datasets and demonstrate its flexibility with tracking results and sequences from the Visual Object Tracking (VOT) challenge.

Session

Recognition, Optimisation and Performance Evaluation

Files

Extended Abstract (PDF, 516K)
Paper (PDF, 6M)

DOI

10.5244/C.30.140
https://dx.doi.org/10.5244/C.30.140

Citation

Bibtex

        @inproceedings{BMVC2016_140,
        	title={Detecting tracking errors via forecasting},
        	author={Obaidullah Khalid, Andrea Cavallaro and Bernhard Rinner},
        	year={2016},
        	month={September},
        	pages={140.1-140.12},
        	articleno={140},
        	numpages={12},
        	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
        	publisher={BMVA Press},
        	editor={Richard C. Wilson, Edwin R. Hancock and William A. P. Smith},
        	doi={10.5244/C.30.140},
        	isbn={1-901725-59-6},
        	url={https://dx.doi.org/10.5244/C.30.140}
        }