Title :
Variable step search algorithm for motion estimation
Author :
Cai, Z.Q. ; Tran, V.N.
Author_Institution :
Dept. of Commun. & Electr. Eng., R. Melbourne Inst. of Technol., Vic., Australia
Abstract :
Motion estimation (ME) is a key technique in interframe coding. It is the basis of most compression algorithms for video compression, such as the CCITT standard H. 261, and MPEG 2. The performance of ME is decided by two factors: (1) the estimation exactitude; and (2) the computational load. The full search algorithm is the optimal one for the first factor, but it requires extensive computations. To reduce the computational complexity, many efficient search algorithms have been proposed. The one step at a time search (OSATS) is the second most efficient algorithm, but it becomes inefficient when the search window is greater than 4 pels/frame. The aim of this paper is to overcome this disadvantage
Keywords :
computational complexity; data compression; image coding; motion estimation; search problems; CCITT standard H. 261; MPEG 2; compression algorithms; computational complexity reduction; computational load; image coding; interframe coding; motion estimation; one step at a time search; search algorithms; search window; variable step search algorithm; video compression; Australia; Compression algorithms; Computational complexity; Computational modeling; Displacement measurement; Measurement standards; Motion estimation; Variable structure systems; Video coding; Video compression;
Conference_Titel :
Information Theory, 1995. Proceedings., 1995 IEEE International Symposium on
Conference_Location :
Whistler, BC
Print_ISBN :
0-7803-2453-6
DOI :
10.1109/ISIT.1995.550355
