Title:	EXPLOITING SPATIO-TEMPORAL COHERENCY IN TIME-VARYING VISUALIZATION OVER NETWORK ENVIRONMENTS
Author(s):	Lazaro Campoalegre, Tom Noonan and John Dingliana
ISBN:	978-989-8533-66-1
Editors:	Yingcai Xiao and Ajith P. Abraham
Year:	2017
Edition:	Single
Keywords:	Time-varying data, Volume Rendering
Type:	Full Paper
First Page:	163
Last Page:	172
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	We propose an efficient client-server interaction system for multi-user visualization of real-time time-varying volumetric simulations. The approach is designed to reduce data transfer over network environments, while enabling users to interact independently with the visualization. Exploiting spatio-temporal coherence and parallel processing on Graphical Processing Units (GPU's) on both server and client, a predictive mechanism on the server evaluates, on the fly, whether blocks of data can be extrapolated with sufficient accuracy based on previous frames of simulation. If so, a simple instruction is sent to the client to extrapolate the block, in lieu of transmitting actual volume data of the block. A comparative analysis indicates that we can achieve visualization performance, for time-varying datasets, on par with previous techniques whilst considerably reducing the size of data sent through the network. Perceptual evaluations using an established metric indicate that the visual quality is not noticeably reduced.
	Go Back