Online selection of discriminative tracking features

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 27; no. 10; pp. 1631 - 1643
• Main Authors: Collins, R.T., Yanxi Liu, Leordeanu, M.
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01.10.2005; IEEE Computer Society; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Artificial Intelligence; Cameras; Clutter; Computer science; control theory; systems; Computer Systems; Density; Discriminant Analysis; Exact sciences and technology; feature creation; feature evaluation and selection; Head; Histograms; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Index Terms- Computer vision; Information Storage and Retrieval - methods; Layout; Likelihood ratio; Motion; Motion detection; Movement; On-line systems; Online; Online Systems; Pattern recognition; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; Robustness; Skin; Studies; Subtraction Technique; Target tracking; Tasks; time-varying imagery; Tracking; Variance; Video Recording - methods; Time varying system; Computer vision; feature creation; Target tracking; feature evaluation and selection; Likelihood ratio; time-varying imagery; Pattern analysis; tracking
• ISSN: 0162-8828; 1939-3539
• DOI: 10.1109/TPAMI.2005.205

This paper presents an online feature selection mechanism for evaluating multiple features while tracking and adjusting the set of features used to improve tracking performance. Our hypothesis is that the features that best discriminate between object and background are also best for tracking the object. Given a set of seed features, we compute log likelihood ratios of class conditional sample densities from object and background to form a new set of candidate features tailored to the local object/background discrimination task. The two-class variance ratio is used to rank these new features according to how well they separate sample distributions of object and background pixels. This feature evaluation mechanism is embedded in a mean-shift tracking system that adaptively selects the top-ranked discriminative features for tracking. Examples are presented that demonstrate how this method adapts to changing appearances of both tracked object and scene background. We note susceptibility of the variance ratio feature selection method to distraction by spatially correlated background clutter and develop an additional approach that seeks to minimize the likelihood of distraction.