Robot vision
Signal Processing for Computer Vision
Signal Processing for Computer Vision
Image Features Based on a New Approach to 2D Rotation Invariant Quadrature Filters
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
Representing local structure using tensors II
SCIA'11 Proceedings of the 17th Scandinavian conference on Image analysis
Optimization of quadrature filters based on the numerical integration of improper integrals
DAGM'11 Proceedings of the 33rd international conference on Pattern recognition
A study of the yosemite sequence used as a test sequence for estimation of optical flow
SCIA'05 Proceedings of the 14th Scandinavian conference on Image Analysis
| Hi-index | 0.00 |
The question of which properties of a local structure estimator are important is discussed. Answers are provided via the introduction of a number of fundamental invariances. Mathematical formulations corresponding to the required invariances leads up to the introduction of a new class of filter sets termed loglets. Using loglets it is shown how the concepts of quadrature and phase can be defined in n-dimensions. A number of experiments support the claim that loglets are preferable to other designs. In particular it is demonstrated that the loglet approach outperforms a Gaussian derivative approach in resolution and robustness to variations in object illumination. It is also shown how a measure of the certainty of the estimate can be obtained using the consistency of the generalized phase with respect to orientation.