Stereoscopic tracking of bodies in motion
Image and Vision Computing - Special issue: 5th Alvey vision meeting
A head-eye system—analysis and design
CVGIP: Image Understanding - Special issue on purposive, qualitative, active vision
A framework for spatiotemporal control in the tracking of visual contours
International Journal of Computer Vision
Evaluation of Interest Point Detectors
International Journal of Computer Vision - Special issue on a special section on visual surveillance
On sequential Monte Carlo sampling methods for Bayesian filtering
Statistics and Computing
Color-Based Probabilistic Tracking
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
IEEE Transactions on Pattern Analysis and Machine Intelligence
Scale & Affine Invariant Interest Point Detectors
International Journal of Computer Vision
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Probabilistic Object Tracking Using Multiple Features
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 2 - Volume 02
Object tracking in image sequences using point features
Pattern Recognition
Real-time and robust monocular SLAM using predictive multi-resolution descriptors
ISVC'06 Proceedings of the Second international conference on Advances in Visual Computing - Volume Part II
SURF: speeded up robust features
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part I
Evaluation of Interest Point Detectors and Feature Descriptors for Visual Tracking
International Journal of Computer Vision
| Hi-index | 0.00 |
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-resolution framework, giving a representation of those points at different scales. Inspired by [1], a stack of descriptors is built only the first time that the interest points are detected and extracted from the region of interest. This provides efficiency of representation 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 and the scale obtained by the UKF, the search of the object can be constrained only in a specific region in both the image and in scale. The use of this approach shows an improvement in real-time tracking and in the ability to recover from full occlusions.