Robust Feature Descriptors for Efficient Vision-Based Tracking
G. Carrera, J. Savage, and W. Mayol-Cuevas. Progress in Pattern Recognition, Image Analysis and Applications, (2008)Real time (10 fps) tracking using the SURF features..
Abstract
This paper presents a robust implementation of an object tracker able to tolerate partial occlusions, rotation and scale for
a variety of different objects. The objects are represented by collections of interest points which are described in a multi-resolutionframework, giving a representation of those points at different scales. Inspired by 1, a stack of descriptors is built onlythe first time that the interest points are detected and extracted from the region of interest. This provides efficiency ofrepresentation and results in faster tracking due to the fact that it can be done off-line. An Unscented Kalman Filter (UKF)using a constant velocity model estimates the position and the scale of the object, with the uncertainty in the position andthe scale obtained by the UKF, the search of the object can be constrained only in a specific region in both the image andin scale.
%0 Journal Article
%1 Carrera2008
%A Carrera, Gerardo
%A Savage, Jesus
%A Mayol-Cuevas, Walterio
%D 2008
%J Progress in Pattern Recognition, Image Analysis and Applications
%K 3d accuracy machinevision objecttracking surf tracking
%P 251--260
%T Robust Feature Descriptors for Efficient Vision-Based Tracking
%U http://dx.doi.org/10.1007/978-3-540-76725-1_27
%X This paper presents a robust implementation of an object tracker able to tolerate partial occlusions, rotation and scale for
a variety of different objects. The objects are represented by collections of interest points which are described in a multi-resolutionframework, giving a representation of those points at different scales. Inspired by 1, a stack of descriptors is built onlythe first time that the interest points are detected and extracted from the region of interest. This provides efficiency ofrepresentation and results in faster tracking due to the fact that it can be done off-line. An Unscented Kalman Filter (UKF)using a constant velocity model estimates the position and the scale of the object, with the uncertainty in the position andthe scale obtained by the UKF, the search of the object can be constrained only in a specific region in both the image andin scale.
@article{Carrera2008,
abstract = {This paper presents a robust implementation of an object tracker able to tolerate partial occlusions, rotation and scale for
a variety of different objects. The objects are represented by collections of interest points which are described in a multi-resolutionframework, giving a representation of those points at different scales. Inspired by [1], a stack of descriptors is built onlythe first time that the interest points are detected and extracted from the region of interest. This provides efficiency ofrepresentation and results in faster tracking due to the fact that it can be done off-line. An Unscented Kalman Filter (UKF)using a constant velocity model estimates the position and the scale of the object, with the uncertainty in the position andthe scale obtained by the UKF, the search of the object can be constrained only in a specific region in both the image andin scale.},
added-at = {2009-06-04T09:34:25.000+0200},
author = {Carrera, Gerardo and Savage, Jesus and Mayol-Cuevas, Walterio},
biburl = {https://www.bibsonomy.org/bibtex/2ede17851029e3b9383f76bbd1f29fb63/midtiby},
description = {SpringerLink - Book Chapter},
interhash = {ebe25e21990e976cfca92a97254c9ce2},
intrahash = {ede17851029e3b9383f76bbd1f29fb63},
journal = {Progress in Pattern Recognition, Image Analysis and Applications},
keywords = {3d accuracy machinevision objecttracking surf tracking},
note = {Real time (10 fps) tracking using the SURF features.},
pages = {251--260},
timestamp = {2009-06-04T09:34:25.000+0200},
title = {Robust Feature Descriptors for Efficient Vision-Based Tracking},
url = {http://dx.doi.org/10.1007/978-3-540-76725-1_27},
year = 2008
}