Tracking using multilevel quantizations

Publication Type:
Conference Proceeding
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2014, 8694 LNCS (PART 6), pp. 155 - 171
Issue Date:
Full metadata record
Most object tracking methods only exploit a single quantization of an image space: pixels, superpixels, or bounding boxes, each of which has advantages and disadvantages. It is highly unlikely that a common optimal quantization level, suitable for tracking all objects in all environments, exists. We therefore propose a hierarchical appearance representation model for tracking, based on a graphical model that exploits shared information across multiple quantization levels. The tracker aims to find the most possible position of the target by jointly classifying the pixels and superpixels and obtaining the best configuration across all levels. The motion of the bounding box is taken into consideration, while Online Random Forests are used to provide pixel- and superpixel-level quantizations and progressively updated on-the-fly. By appropriately considering the multilevel quantizations, our tracker exhibits not only excellent performance in non-rigid object deformation handling, but also its robustness to occlusions. A quantitative evaluation is conducted on two benchmark datasets: a non-rigid object tracking dataset (11 sequences) and the CVPR2013 tracking benchmark (50 sequences). Experimental results show that our tracker overcomes various tracking challenges and is superior to a number of other popular tracking methods. © 2014 Springer International Publishing.
Please use this identifier to cite or link to this item: