Abstract
Abstract Multiple objects tracking is a challenging task. This article presents an algorithm which can detect and track multiple objects, and update target model automatically. The contributions of this paper as follow: Firstly,we also use color histogram(CH) and histogram of orientated gradients(HOG) to represent the objects, model update is realized by kalman filter and gaussian model; secondly we use Gaussian Mixture Model(GMM) and Bhattacharyya distance to detect object appearance. Particle filter with combined features and model update mechanism can improve tracking results. Experiments on video sequences demonstrate that the method presented in this paper can realize multiple objects detection and tracking.
Highlights
Visual object tracking is an important task in the field of computer vision
This paper presents a novel particle filter for tracking multiple objects
In process of tracking the features of the object may slow change so we introduce the kalman filter and GMM to do the features adaptive update
Summary
Visual object tracking is an important task in the field of computer vision. It is widely used in motionbased recognition[1], automated surveillance, humancomputer interaction[2], vehicle navigation and traffic monitoring[3]. CH together with HOG can be applied to deal with some appearance and shape changes to achieve robust tracking performance, model update should be taken into consideration. Hanzi Wang et al.[14] proposed a similarity measure based on Spatial-color mixture of Gaussians appearance model for particle filters. Katja Nummiaro et al.[16] presented the integration of color distributions into particle filter, which has been typically used in combination with edge-based image features. Budi Sugandi et al.[19] proposed a method for multiple object tracking based on color features. Not all of the tracking system based on color particle filter employe model update. Some of color particle filter introduce model update as Ref. 19, but this type of methods are subject to tracking drift.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have