Title :
Vector tracing techniques for motion estimation algorithms in video coding
Author :
Mattavelli, Marco ; Zoia, Giorgio
Author_Institution :
Integrated Syst. Center - C3i, Swiss Fed. Inst. of Technol., Lausanne, Switzerland
Abstract :
Despite the several efforts aiming at reducing the complexity, block motion estimation remains the most computationally demanding stage of video compression algorithms. This is particularly evident when sequences contain large displacements. Very large search windows are needed to achieve high quality coding when such critical conditions occur. This paper presents a block motion estimation technique based on the combination of motion trajectories tracing and of a modified genetic search heuristic. The proposed method is able to provide motion estimates in very large search windows with optimal coding results. The complexity reduction factor ranges up to more than two orders of magnitudes. The technique can be applied to any macroblock-based video compression standard and to any group of picture (GOP) structure.
Keywords :
computational complexity; genetic algorithms; motion estimation; search problems; video coding; block motion estimation algorithm; complexity reduction factor; group of picture structure; modified genetic search heuristic; motion trajectories tracing; optimal coding; vector tracing technique; video coding; video compression algorithm; Complexity theory; Encoding; Estimation; Genetics; Motion estimation; PSNR; Sociology;
Conference_Titel :
Signal Processing Conference (EUSIPCO 1998), 9th European
Conference_Location :
Rhodes
Print_ISBN :
978-960-7620-06-4