Combining multiple scoring systems for target tracking using rank–score characteristics

Abstract

Abstract

Comparing discrimination and CFA for selecting tracking features

“…In [10], S consisted of average rank and Mahalanobis score combinations of the three feature measurements: averaged target color, target bounding-box ratio and target region centroid location. In [9], S consisted of the 11 fusion operations consisting of all combinations of 2 and 3 features fused by average rank combination or Mahalanobis score combination of these same three features.…”

“…In [9], S consisted of the 11 fusion operations consisting of all combinations of 2 and 3 features fused by average rank combination or Mahalanobis score combination of these same three features. Both [9] and [10] use ground-truth experiments to show that a diversity measure calculated from the rank-score functions of each feature can be used to determine which of the fusion operations to apply to generate improved performance. …”

“…We showed that this diversity could be used to select between rank and score fusions of all three features [10] or, in combination with a relative performance measure [9] to select which pair of three features in addition to rank versus score fusion, for improved tracking accuracy. However, relative performance is only available as a metric in groundtruth experiments, and hence can not be used as part of a general purpose video tracker.…”

“…In [10] it was shown that a smaller value of rank/score diversity indicated a score fusion produced a better performance, while a larger value indicated than a rank fusion produced a better result.…”

“…Interactions between target and background include targets being occluded by the background and targets appearing to merge into the background due to lighting changes or changes in the background scene. In this paper, we present a novel solution to these two issues based on integrating two existing approaches: combinatorial fusion analysis [6] applied to video target tracking using a rank-score diversity measure [10] developed previously by the authors, and a target background/foreground discrimination measure [2] developed by Collins & Liu. The discrimination approach was developed by Collins & Liu in a framework very similar to combinatorial fusion but was applied by them to single target tracking.…”