Adaptive Kernel Based Tracking Using Mean-Shift

@inproceedings{Pu2006AdaptiveKB,
  title={Adaptive Kernel Based Tracking Using Mean-Shift},
  author={Jie-Xin Pu and Ningsong Peng},
  booktitle={ICIAR},
  year={2006}
}
The mean shift algorithm is an kernel based way for efficient object tracking. However, there is presently no clean mechanism for selecting kernel bandwidth when the object size is changing. We present an adaptive kernel bandwidth selection method for rigid object tracking. The kernel bandwidth is updated by using the object affine model that is estimated by using object corner correspondences between two consecutive frames. The centroid of object is registered by a special backward tracking… 
Target Tracking Based on Mean Shift and KALMAN Filter with Kernel Histogram Filtering
TLDR
KALMAN filter is a suitable approach to handle model update to handle mean shift algorithm with model update ability for tracking and achieve good results.
Robust scale-adaptive mean-shift for tracking
TLDR
A novel histogram color weighting that exploits the object neighborhood to help discriminate the target called background ratio weighting (BRW) is presented and it is shown that the BRW improves performance of MS-like tracking methods in general.
Robust Scale-Adaptive Mean-Shift for Tracking
TLDR
This paper addresses the problem of scale adaptation of the Hellinger distance based Mean-Shift tracker with a simple and scalable implementation.
Short-Term Visual Object Tracking in Real-Time
TLDR
A novel tracking framework HMMTxD that fuses multiple tracking methods together with a proposed feature-based online detector and a framework for fusion of multiple trackers and detector and contributions to the problem of tracker evaluation within the Visual Object Tracking (VOT) initiative are proposed.
Bandwidth Selection for Mean-shift based Unsupervised Learning Techniques: a Unified Approach via Self-coverage
TLDR
This paper proposes to use a so-called self-coverage measure as a general device for bandwidth selection in this context and shows how a bandwidth h will be favorable if a high proportion of data points falls within circles or ``hypertubes"; of radius h centered at the fitted object.
A trajectory-based ball detection and tracking system with applications to shooting angle and velocity estimation in basketball videos
TLDR
A system to automatically analyse a basketball long-shot using trajectory-based ball tracking method from a basketball video sequence that minimised the rate of error of ball tracking in the video and showed encouraging results for videos with dynamic background and different illumination conditions.
Comparative Estimation of Trajectory based Tracking System and Impact of Subsequent on Projectile Course in 3-D Utilizing Motion Technique
There are various approaches to estimate the trajectory based tracking system and subsequent brunt point of a projectile course. The mainly methods are more accurate and require a lot of input data,

References

SHOWING 1-10 OF 12 REFERENCES
Automatic Selection of Kernel-Bandwidth for Mean-Shift Object Tracking
TLDR
An automatic bandwidth selection method is proposed based on backward tracking and object centroid registration based on the analysis of similarity of object kernel-histogram in different scales, i.e. the Bhattacharyya coefficient, and is applied to track vehicle changing in size with encouraging results.
Kernel-Based Object Tracking
TLDR
A new approach toward target representation and localization, the central component in visual tracking of nonrigid objects, is proposed, which employs a metric derived from the Bhattacharyya coefficient as similarity measure, and uses the mean shift procedure to perform the optimization.
Tracking of object with SVM regression
TLDR
This paper presents a novel feature-matching based approach for rigid object tracking that can also be used for general feature-based epipolar constraint estimation, in which it can quickly detect outliers even if they make up, in theory, over 50% of the whole data.
Mean-shift blob tracking through scale space
  • R. Collins
  • Mathematics, Computer Science
    2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings.
  • 2003
TLDR
Lindeberg's theory of feature scale selection based on local maxima of differential scale-space filters to the problem of selecting kernel scale for mean-shift blob tracking shows that a difference of Gaussian (DOG) mean- shift kernel enables efficient tracking of blobs through scale space.
Real-time tracking of non-rigid objects using mean shift
TLDR
The theoretical analysis of the approach shows that it relates to the Bayesian framework while providing a practical, fast and efficient solution for real time tracking of non-rigid objects seen from a moving camera.
Target tracking in airborne forward looking infrared imagery
TLDR
A robust approach for tracking targets in forward looking infrared (FLIR) imagery taken from an airborne moving platform, which combines automatic model update and global motion compensation into one framework is proposed.
Maximum-likelihood template matching
  • C. Olson
  • Mathematics, Computer Science
    Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662)
  • 2000
TLDR
A robust measure and efficient search strategy for template matching with a binary or greyscale template using a maximum-likelihood formulation and the use of these techniques for object recognition, stereo matching, feature selection, and tracking is examined.
Mean Shift, Mode Seeking, and Clustering
  • Yizong Cheng
  • Mathematics, Computer Science
    IEEE Trans. Pattern Anal. Mach. Intell.
  • 1995
TLDR
Mean shift, a simple interactive procedure that shifts each data point to the average of data points in its neighborhood is generalized and analyzed and makes some k-means like clustering algorithms its special cases.
Robust window operators
TLDR
A robust window operator is demonstrated that preserves gray-level and gradient discontinuities in digital images as it smooths and estimates derivatives.
The estimation of the gradient of a density function, with applications in pattern recognition
TLDR
Applications of gradient estimation to pattern recognition are presented using clustering and intrinsic dimensionality problems, with the ultimate goal of providing further understanding of these problems in terms of density gradients.
...
1
2
...