Color-Based Probabilistic Tracking

Abstract

Abstract. Color-based trackers recently proposed in [3,4,5] have been proved robust and versatile for a modest computational cost. They are especially appealing for tracking tasks where the spatial structure of the tracked objects exhibits such a dramatic variability that trackers based on a space-dependent appearance reference would break down very fast. Trackers in [3,4,5] rely on the deterministic search of a window whose color content matches a reference histogram color model. Relying on the same principle of color histogram distance, but within a probabilistic framework, we introduce a new Monte Carlo tracking technique. The use of a particle filter allows us to better handle color clutter in the background, as well as complete occlusion of the tracked entities over a few frames. This probabilistic approach is very flexible and can be extended in a number of useful ways. In particular, we introduce the following ingredients: multi-part color modeling to capture a rough spatial layout ignored by global histograms, incorporation of a background color model when relevant, and extension to multiple objects.

Multi-view Object Tracking Using Sequential Belief Propagation

“…Thus, the observation process is to match the color histogram in a candidate region, a particle, with a pre-learned reference model, where the Bhattacharyya similarity coefficient is computed to measure the distance. The effectiveness of this model has been shown previously [11,12,4] and is confirmed by this work. In all the experiments, we manually initialize the regions of targets of interest at the first frame of each camera and learn the reference color models.…”

“…Following Perez et al [11], a classical color observation model based on HSV color histograms is adopted which has the advantage of being insensitive to illumination effects. Thus, the observation process is to match the color histogram in a candidate region, a particle, with a pre-learned reference model, where the Bhattacharyya similarity coefficient is computed to measure the distance.…”

Multi-step Multi-camera View Planning for Real-Time Visual Object Tracking

“…The cameras are calibrated to a global coordinate system with the calibration pattern, which is not used for the tracking process. The object is tracked with a color histogram tracker [14].…”

Multi-Target Tracking Using Color Information

“…Likewise, Limin [20] uses color-based Kalman particle filter object tracking algorithm. In crowded environments, a hue-saturation histogram with a particle filter based probabilistic technique is implemented in [21] for tracking objects.…”