Determining the Epipolar Geometry and its Uncertainty: A Review
International Journal of Computer Vision
Efficient Region Tracking With Parametric Models of Geometry and Illumination
IEEE Transactions on Pattern Analysis and Machine Intelligence
Kalman Filtering and Neural Networks
Kalman Filtering and Neural Networks
Factorization with Uncertainty
ECCV '00 Proceedings of the 6th European Conference on Computer Vision-Part I
Bundle Adjustment - A Modern Synthesis
ICCV '99 Proceedings of the International Workshop on Vision Algorithms: Theory and Practice
Robust detection of degenerate configurations for the fundamental matrix
ICCV '95 Proceedings of the Fifth International Conference on Computer Vision
A Unified Factorization Algorithm for Points, Line Segments and Planes with Uncertainty Models
ICCV '98 Proceedings of the Sixth International Conference on Computer Vision
An Invitation to 3-D Vision: From Images to Geometric Models
An Invitation to 3-D Vision: From Images to Geometric Models
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Object Tracking: Feature Selection and Confidence Propagation
CRV '04 Proceedings of the 1st Canadian Conference on Computer and Robot Vision
Feature Uncertainty Arising from Covariant Image Noise
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Learning Bayesian Networks
An iterative image registration technique with an application to stereo vision
IJCAI'81 Proceedings of the 7th international joint conference on Artificial intelligence - Volume 2
Error concealment via Kalman filter for heavily corrupted videos in H.264/AVC
Image Communication
| Hi-index | 0.01 |
Accurate feature tracking is the foundation of many high level tasks in computer vision, such as 3D reconstruction and motion analysis. Although there are many feature tracking algorithms, most of them do not maintain information about the error of the data being tracked. Also, due to the difficulty and spatial locality of the problem, existing methods can generate grossly incorrect correspondences, making outlier rejection an essential post-processing step. We propose a new generic framework that uses the Scaled Unscented Transform to augment arbitrary feature tracking algorithms, and use Gaussian Random Variables (GRV) for the representation of features' locations uncertainties. We apply and validate the framework on the well-understood Kanade-Lucas-Tomasi feature tracker, and call it Unscented KLT (UKLT). The UKLT tracks GRVs and rejects incorrect correspondences, without a global model of motion. We validate our method on real and synthetic sequences, and demonstrate how the UKLT outperforms other approaches on both outlier rejection and the accuracy of feature locations.