Title :
Mutual Information for Lucas-Kanade Tracking (MILK): An Inverse Compositional Formulation
Author :
Dowson, Nicholas ; Bowden, Richard
Author_Institution :
Siemens Molecular Imaging, Oxford
Abstract :
Mutual information (Ml) is popular for registration via function optimization. This work proposes an inverse compositional formulation of Ml for Levenberg-Marquardt optimization. This yields a constant Hessian, which may be precomputed. Speed improvements of 15 percent were obtained, with convergence accuracies similar those of the standard formulation.
Keywords :
convergence; functions; image registration; inverse problems; optical tracking; optimisation; Levenberg-Marquardt optimization; Lucas-Kanade tracking; constant Hessian; convergence; function optimization; image registration; inverse compositional formulation; mutual information; Computer vision; Image Processing and Computer Vision; Optimization; Tracking; Algorithms; Artificial Intelligence; Image Enhancement; Image Interpretation, Computer-Assisted; Imaging, Three-Dimensional; Motion; Pattern Recognition, Automated; Subtraction Technique;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2007.70757
