Cookies

We use cookies to ensure that we give you the best experience on our website. By continuing to browse this repository, you give consent for essential cookies to be used. You can read more about our Privacy and Cookie Policy.


Durham Research Online
You are in:

Real-time construction and visualisation of drift-free video mosaics from unconstrained camera motion.

Breszcz, M. and Breckon, T.P. (2015) 'Real-time construction and visualisation of drift-free video mosaics from unconstrained camera motion.', Journal of engineering. . pp. 1-12.

Abstract

This work proposes a novel approach for real-time video mosaicking facilitating drift-free mosaic construction and visualisation, with integrated frame blending and redundancy management, that is shown to be flexible to a range of varying mosaic scenarios. The approach supports unconstrained camera motion with in-sequence loop closing, variation in camera focal distance (zoom) and recovery from video sequence breaks. Real-time performance, over extended duration sequences, is realised via novel aspects of frame management within the mosaic representation and thus avoiding the high data redundancy associated with temporally dense, spatially overlapping video frame inputs. This managed set of image frames is visualised in real time using a dynamic mosaic representation of overlapping textured graphics primitives in place of the traditional globally constructed, and hence frequently reconstructed, mosaic image. Within this formulation, subsequent optimisation occurring during online construction can thus efficiency adjust relative frame positions via simple primitive position transforms. Effective visualisation is similarly facilitated by online inter-frame blending to overcome the illumination and colour variance associated with modern camera hardware. The evaluation illustrates overall robustness in video mosaic construction under a diverse range of conditions including indoor and outdoor environments, varying illumination and presence of in-scene motion on varying computational platforms.

Item Type:Article
Full text:(VoR) Version of Record
Available under License - Creative Commons Attribution.
Download PDF
(3732Kb)
Status:Peer-reviewed
Publisher Web site:http://dx.doi.org/10.1049/joe.2015.0016
Publisher statement:This is an open access article published by the IET under the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/)
Date accepted:18 June 2015
Date deposited:05 October 2015
Date of first online publication:August 2015
Date first made open access:No date available

Save or Share this output

Export:
Export
Look up in GoogleScholar