Title :
An Efficient Data Reuse Motion Estimation Engine
Author :
Goel, Sumeer ; Ismail, Yasser ; Devulapalli, Parimal ; McNeely, Jason ; Bayoumi, Magdy A.
Author_Institution :
Centre for Adv. Comput. Studies, Louisiana Univ., Lafayette, LA
Abstract :
We present a novel motion estimation engine (MEE) architecture that efficiently reuses search area data while fully utilizing the hardware resources. A 2-D processing element (PE) core is central to the architecture. Search area data flows both horizontally as well as vertically while the current block data is stationary. A clever PE design ensures simple but highly regular dataflow through the core avoiding long interconnect delays. For a search range of [-16,+15] and block size of 16, our architecture can perform motion estimation for 60 fps of 4CIF video at 100 MHz
Keywords :
motion estimation; search engines; video signal processing; 100 MHz; 2D processing element core; 4CIF video; data reuse MEE architecture; motion estimation engine; search area data; Automatic voltage control; Bandwidth; Computer architecture; Delay; Engines; Hardware; Motion estimation; Systolic arrays; Transform coding; Video coding;
Conference_Titel :
Signal Processing Systems Design and Implementation, 2006. SIPS '06. IEEE Workshop on
Conference_Location :
Banff, Alta.
Print_ISBN :
1-4244-0383-9
Electronic_ISBN :
1520-6130
DOI :
10.1109/SIPS.2006.352613