Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/1165
Title: An efficient bidirectional frame prediction using particle swarm optimization technique
Authors: Ranganadham, D
Gorpuni, P K
Keywords: Average mean square prediction,
Macroblock,
Particle swarm optimization,
Motion vector.
Issue Date: 2009
Citation: International Conference on Advances in Recent Technologies in Communication and Computing, 2009, Article number 5328092, Pages 42-46
Abstract: This paper presents a Novel Bidirectional motion estimation technique, which is based on the Particle swarm optimization algorithm. Particle swarm optimization (PSO) is a population based optimization technique which has the potentiality to avoid local minima solution which is usually encountered by the traditional block matching algorithms (BMA) such as the three step search (TSS) and the diamond search (DS). To speed up the search, static macro blocks are found in our method, which is particularly beneficial to those video sequences containing small motion contents. Skipping such static macro blocks from processing can save the computation time and memory also. In the proposed method each time we are finding the best matching macro block in two frames at a time so we can reduce the number of error function calculations and it is faster than if we apply PSO technique to find forward motion vector and backward motion vector separately. The proposed method is applied to a number of benchmark video sequences and the results are compared with those obtained by applying the existing methods. Simulation results shows that the proposed algorithms gives the close match of PSNR values when compared to joint search algorithm with DS. Thus, PSO algorithm for Bidirectional motion estimation is empirically given to reduce computational complexity.
URI: http://dx.doi.org/10.1109/ARTCom.2009.87
http://hdl.handle.net/2080/1165
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
Published_paper.pdf475.8 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.