A Highly Efficient Real-Time Tracking Method in Augmented Reality System

Abstract

A highly efficient real-time tracking algorithm is very important for augmented reality implementation by computer vision mode. The mean-shift (MS) algorithm is widely used in object tracking for augmented reality system because of its not only speed but also highly efficient and real-time features. The traditional MS algorithm only use color feature as one of the important cues in image sequences. Even though the MS tracking algorithm cannot very suitable for complex environments, such as the background with object’s similar color, sudden light changes, occlusion types and so on. The proposed algorithm use a convex kernel function in association with the motion information to improve the MS tracking algorithm for the purpose to solve the above problems. Using these new features fusion, a robust mean-shift tracking algorithm is proposed. Experimental work show the proposed mean-shift algorithm has an optimum performance in robust and realtime object tracking for the augmented reality system.

Authors and Affiliations

Wei-Bin Chen, Ben Niu, Hong-Bin Gu

Keywords

Related Articles

Efficacy of Facts in Power Oscillation Damping and Renewable Integration

This paper discusses about the approach towards effective power oscillation damping, root causes for the genesis of power oscillation and various methodologies adopted to eradicate it.This paper presents an unified appro...

Corner Cut Microstrip Patch Antenna for Ultra Wide BandApplications

In this paper, the radiation performance of a small printedmicro strip antennadesigned on glass epoxy FR4 substrate discussed. The proposed antenna is capable to cover Wi MAX, Wi Fi, WBAN and Bluetooth operations and UWB...

Impact of Large Scale Renewable Integration in Bangladesh Power System

Bangladesh, a small developing country of south Asia observing a rapid economic transition from agriculture to industry and facing a threat of unsatisfied consumer demand of energy exacerbating with growing population. K...

A Review of Maximum Power Point Tracking Algorithm for Solar Photovoltaic Applications

The world’s large dependency on conventional energy sources has not only posed a threat to the environment but also they are non-renewable. Therefore, a huge interest is put upon renewable sources of energy. Amongst them...

Comparitive Analysis of Fuzzy Based MPPT for Boost and SEPIC Converter Topologies for PV Application

This paper proposes the Maximum Power Point Tracking based on Fuzzy logic control for PV System.MPPT is required to extract maximum power corresponding to specific operating point. The output power of PV System is depend...

Download PDF file
  • EP ID EP388675
  • DOI 10.9790/1676-1301045460.
  • Views 159
  • Downloads 0

How To Cite

Wei-Bin Chen, Ben Niu, Hong-Bin Gu (2018). A Highly Efficient Real-Time Tracking Method in Augmented Reality System. IOSR Journals (IOSR Journal of Electrical and Electronics Engineering), 13(1), 54-60. https://www.europub.co.uk/articles/-A-388675