Finite asymmetric generalized Gaussian mixture models learning for infrared object detection.

Highlights: [•] We develop a method for learning the parameters of multidimensional asymmetric generalized Gaussian mixtures (AGGM). [•] We describe and illustrate an online algorithm, based on the developed AGGM, for pedestrian detection using infrared images. [•] We implement a multiple-target tracking (MTT) framework using AGGM. [•] We demonstrate the importance of the fusion of both visible and thermal images for MTT. [ABSTRACT FROM AUTHOR]