F. Bunyak, and
Proc. IEEE Applied Imagery Pattern Recognition (AIPR),
Robust and accurate visual tracking is needed for many computer vision
applications from video summarization to visual surveillance. Visual
tracking remains to be a challenging task because of factors such as changing
object appearance, illumination variations and shadows, partial and
full occlusions, camera motion, distractors, and scale changes. Recently
our group proposed a Likelihood of Features Tracking (LoFT) system
that fuses multiple sources of information about target and its environment
to perform robust single object tracking. LoFT has been shown to
successfully track objects under different scenarios from full motion video
to wide-are motion imagery.
In this paper, we extend the LoFT framework with color information
and multi-scale feature sets to boost its performance. Color information
is incorporated through Color Name (CN) mapping scheme.Multiscale
selective has been done according to select the suitable scale among the
variation of scale set.
Experiments on VOT 2015 tracking benchmark, which includes video
sequences with significant color and scale variations, show significant performance
improvement over baseline LoFT both in accuracy and robustness
11% and 24% respectively.
author = "N. Al-Shakarji and F. Bunyak and K. Palaniappan",
title = "CS-LOFT - Color and scale adaptive tracking using Bhattacharyya distance and max pooling consistency",
year = 2016,
journal = "Proc. IEEE Applied Imagery Pattern Recognition (AIPR)",
month = "Oct",
keywords = "color, wami, tracking, fmv, features"
N. Al-Shakarji, F. Bunyak, and K. Palaniappan. CS-LOFT - Color and scale adaptive tracking using Bhattacharyya distance and max pooling consistency. Proc. IEEE Applied Imagery Pattern Recognition (AIPR), October 2016.