Title :
How to put probabilities on homographies
Author :
Begelfor, Evgeni ; Werman, Michael
Author_Institution :
Dept. of Comput. Sci., Hebrew Univ., Jerusalem, Israel
Abstract :
We present a family of "normal" distributions over a matrix group together with a simple method for estimating its parameters. In particular, the mean of a set of elements can be calculated. The approach is applied to planar projective homographies, showing that using priors defined in this way improves object recognition.
Keywords :
image recognition; normal distribution; object recognition; parameter estimation; probability; matrix group; normal distribution; object recognition; parameter estimation; planar projective homographies; probabilities; Bayesian methods; Cameras; Covariance matrix; Gaussian distribution; Image recognition; Object recognition; Parameter estimation; Random variables; Statistical distributions; Transmission line matrix methods; Bayesian statistics; Index Terms- Homography; geodesics.; lie groups; normal distribution; Algorithms; Artificial Intelligence; Bayes Theorem; Computer Simulation; Data Interpretation, Statistical; Image Enhancement; Image Interpretation, Computer-Assisted; Information Storage and Retrieval; Models, Statistical; Pattern Recognition, Automated; Statistical Distributions; Subtraction Technique;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2005.200
