Compressive object tracking — A review and analysis

The objective of this article is to audit the tracking strategies, characterize them into distinctive classifications, furthermore distinguish new patterns. To give an improvement to the solution of drift problem in online tracking a separate section called compression tracking has been chosen. Various compressive tracking techniques have been taken along with their working, merits and demerits. Most of the methods include object segmentation using background subtraction. The following methods use diverse strategies like Mean-shift, Kalman filter, Particle filter etc. This paper presents a survey on compressive object tracking using the state of art models used. The designed models which are successfully applied using compressive sensing concepts reduces the number of pixels and these methods are efficient in feature extraction and dimensionality reduction with high accuracy.