Title :
Mean shift is a bound optimization
Author :
Fashing, Mark ; Tomasi, Carlo
Author_Institution :
Dept. of Comput. Sci., Duke Univ., Durham, NC, USA
fDate :
3/1/2005 12:00:00 AM
Abstract :
We build on the current understanding of mean shift as an optimization procedure. We demonstrate that, in the case of piecewise constant kernels, mean shift is equivalent to Newton´s method. Further, we prove that, for all kernels, the mean shift procedure is a quadratic bound maximization.
Keywords :
Newton method; gradient methods; quadratic programming; Newton method; bound optimization; gradient methods; mean shift procedure; piecewise constant kernels; quadratic bound maximization; Application software; Computer vision; Convergence; Density functional theory; Image segmentation; Kernel; Optimization methods; Smoothing methods; Index Terms- Mean shift; Newton´s method; adaptive gradient descent; bound optimization; mode seeking.; Algorithms; Artificial Intelligence; Cluster Analysis; Computer Graphics; Image Enhancement; Image Interpretation, Computer-Assisted; Information Storage and Retrieval; Numerical Analysis, Computer-Assisted; Pattern Recognition, Automated; Sample Size; Signal Processing, Computer-Assisted;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2005.59
