A More Robust Mean Shift Tracker on Joint Color-CLTP Histogram

Abstract

Abstract-A more robust mean shift tracker using the joint of color and Completed Local Ternary Pattern (CLTP) histogram is proposed. CLTP is a generalization of Local Binary Pattern (LBP) which can be applied to obtain texture features that are more discriminant and less sensitive to noise. The joint of color and CLTP histogram based target representation can exploit the target structural information efficiently. To reduce the interference of background in target localization, a corrected background-weighted histogram and background update mechanism are adapted to decrease the weights of both prominent background color and texture features similar to the target object. Comparative experimental results on various challenging videos demonstrate that the proposed tracker performs favorably against several variants of state-of-the-art mean shift tracker when heavy occlusions and complex background changes exist.

Improved Mean Shift Target Localization using True Background Weighted Histogram and Geometric Centroid Adjustment

“…The above methods use only first frame for background weight transformation, whereas in natural videos, the background can change in immediate next frame or after few frames. To improve the MS localization, combination of target structural and spectral information has also been proposed [10]. Local binary pattern based structural feature integration in MS algorithm along spectral information improves target localization [10].…”

“…To improve the MS localization, combination of target structural and spectral information has also been proposed [10]. Local binary pattern based structural feature integration in MS algorithm along spectral information improves target localization [10]. Histogram probabilistic multi hypothesis tracker (H-PMHT) using reciprocal pixel intensity measurement is developed to improve the results in intense clutter environment [17].…”

Efficient Mean-Shift Tracking Using an Improved Weighted Histogram Scheme

“…To handle this problem in object tracking, extensive techniques have been presented. These target representation techniques can be classified into two main categories: pixel-oriented [19][20] and representation-oriented [4,[21][22][23]. Some trackers may adopt the two kinds of techniques at the same time (e.g.…”

“…However, using LBP features in mean-shift tracking remains problematic which requires further investigation. In [22], Pu et al proposed a novel texture descriptor, called the Completed Local Ternary Pattern (CLTP), which is more discriminative and less sensitive to noise in near-uniform image regions such as cheeks and foreheads. They combined CLTP and color into a joint color-CLTP histogram and used the new and distinctive target representation to perform mean-shift tracking.…”

“…Among the various tracking algorithms, mean-shift tracking algorithms have been extensively used in object tracking and have recently been the focus of much research due to their low complexity and robustness [21][22][29][30][31][32]. In this paper, mean-shift is also used to track the target rather than the gradient descent [33][34] method.…”