On the Convergence of the Mean Shift Algorithm in the One-Dimensional Space

Abstract

The mean shift algorithm is a non-parametric and iterative technique that has been used for finding modes of an estimated probability density function. It has been successfully employed in many applications in specific areas of machine vision, pattern recognition, and image processing. Although the mean shift algorithm has been used in many applications, a rigorous proof of its convergence is still missing in the literature. In this paper we address the convergence of the mean shift algorithm in the one-dimensional space and prove that the sequence generated by the mean shift algorithm is a monotone and convergent sequence.

arXiv comment13 pages, 10 figures, Published in Pattern Recognition Letters

Cited by in corpus (7)

• Kernel Smoothing, Mean Shift, and Their Learning Theory with Directional Data
• Measuring Tail Risks
• Nonparametric multimodal regression for circular data
• The EM Perspective of Directional Mean Shift Algorithm
• MeanShift++: Extremely Fast Mode-Seeking With Applications to Segmentation and Object Tracking
• Variational Auto Encoder Gradient Clustering
• A Note on the Convergence of the Gaussian Mean Shift Algorithm