An Improved Mean-Shift Tracker with Kernel Prediction and Scale Optimisation Targeting for Low-Frame-Rate Video Tracking

Z. Li, J. Chen, and N. N. Schraudolph. An Improved Mean-Shift Tracker with Kernel Prediction and Scale Optimisation Targeting for Low-Frame-Rate Video Tracking. In 19^th Intl. Conf. Pattern Recognition (ICPR), Tampa, Florida, 2008.

Download

280.1kB	92.1kB	306.0kB

Abstract

The mean-shift (MS) algorithm is widely used in object tracking because of its speed and simplicity. However, it assumes certain overlap of object appearance and smooth change in object scale between consecutive video frames. This assumption is usually violated in a low-frame-rate (LFR) video, which contains fast motion and scale changes. An LFR video is widely adopted in applications such as surveillance systems, where real-time object tracking is highly desirable but the traditional MS algorithm does not perform well. We addressed this problem by proposing a novel and enhanced mean-shift tracker, named SMDShift, that uses kernel prediction and Stochastic Meta-Descent (SMD) optimization method to deal with the kernel position and scale variation when tracking objects in an LFR video. In our experiments, the SMDShift can track fast moving objects with significant scale change in an LFR video sequence on which the traditional mean-shift and Camshift algorithms fail.

BibTeX Entry

@inproceedings{LiCheSch08,
     author = {Zhidong Li and Jing Chen and Nicol N. Schraudolph},
      title = {\href{http://nic.schraudolph.org/pubs/LiCheSch08.pdf}{
               An Improved Mean-Shift Tracker with Kernel Prediction
               and Scale Optimisation Targeting for Low-Frame-Rate
               Video Tracking}},
  booktitle = {19$^{th}$ Intl.\ Conf.\ Pattern Recognition (ICPR)},
    address = {Tampa, Florida},
       year =  2008,
   b2h_type = {Other},
  b2h_topic = {>Stochastic Meta-Descent, Computer Vision},
   abstract = {
    The mean-shift (MS) algorithm is widely used in object tracking
    because of its speed and simplicity.  However, it assumes certain
    overlap of object appearance and smooth change in object scale
    between consecutive video frames. This assumption is usually
    violated in a low-frame-rate (LFR) video, which contains fast
    motion and scale changes. An LFR video is widely adopted in
    applications such as surveillance systems, where real-time
    object tracking is highly desirable but the traditional MS
    algorithm does not perform well.  We addressed this problem by
    proposing a novel and enhanced mean-shift tracker, named SMDShift,
    that uses kernel prediction and Stochastic Meta-Descent (SMD)
    optimization method to deal with the kernel position and scale
    variation when tracking objects in an LFR video.  In our
    experiments, the SMDShift can track fast moving objects with
    significant scale change in an LFR video sequence on which the
    traditional mean-shift and Camshift algorithms fail.
}}