DocumentCode :
2043876
Title :
Algorithm and architecture co-design of Mixture of Gaussian (MoG) background subtraction for embedded vision
Author :
Tabkhi, Hamed ; Bushey, Robert ; Schirner, Gunar
Author_Institution :
Dept. of Electr. & Comput. Eng., Northeastern Univ., Boston, MA, USA
fYear :
2013
fDate :
3-6 Nov. 2013
Firstpage :
1815
Lastpage :
1820
Abstract :
Embedded vision is a rapidly growing and challenging market that demands high computation with low power consumption. Carefully crafted heterogeneous platforms have the possibility to deliver the required computation within the power budget. However, to achieve efficient realizations, vision algorithms and architectures have to be developed and tuned in conjunction. This article describes the algorithm / architecture co-design opportunities of a Mixture of Gaussian (MoG) implementation for realizing background subtraction. Particularly challenging is the memory bandwidth required for storing the background model (Gaussian parameters). Through joint algorithm tuning and system-level exploration, we develop a compression of Gaussian parameters which allows navigating the bandwidth/quality trade-off. We identify an efficient solution point in which the compression reduces the required memory bandwidth by 63% with limited loss in quality. Subsequently, we propose a HW-based architecture for MoG that includes sufficient flexibility to adjust to scene complexity.
Keywords :
Gaussian processes; computer vision; mixture models; Gaussian parameter compression; HW-based architecture; MoG background subtraction; algorithm tuning; architecture co-design; background model; embedded vision; low power consumption; memory bandwidth; mixture of Gaussian background subtraction; system-level exploration; vision algorithms; Bandwidth; Complexity theory; Computational modeling; Memory management; Software algorithms; Standards;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signals, Systems and Computers, 2013 Asilomar Conference on
Conference_Location :
Pacific Grove, CA
Print_ISBN :
978-1-4799-2388-5
Type :
conf
DOI :
10.1109/ACSSC.2013.6810615
Filename :
6810615
Link To Document :
بازگشت