An Associative Perception-Action Structure Using a Localized Space Variant Information Representation

Weighted Update and Comparison for Channel-Based Distribution Field Tracking

“…Readers unfamiliar with these methods are referred to more comprehensive descriptions in literature [6,2,3] for details.…”

“…Such a tracker, Distribution Field Tracking, DFT, was proposed by Sevilla et al [13] where histograms are used for representing distributions. However, as was shown by Felsberg [4], replacing the histograms with channel representations [6] increases tracker performance, resulting in the Enhanced Distribution Field Tracker, EDFT.…”

Incremental Computation of Feature Hierarchies

“…The channel representation as a computational framework goes back to [10]. In channel representations features are represented by weights assigned to ranges of feature values, similar to histograms but exploiting smooth bins.…”

“…In order to obtain a solution according to (11) with non-negative coefficients, a straightforward idea is to simply set all negative coefficients to zero in each iteration of (10). However, this does not lead to stable results in our experiments.…”

“…Channel representations as a computational framework, e.g. for object recognition, have been introduced in [10], and are directly related to kernel density estimation [11]. Channel representations of features are basically soft-histograms or Parzen estimators with a smooth kernel function.…”