Title :
Graph-based parallelization algorithm for deblocking filter in H.264/AVC
Author :
Seongmin Jo ; Yong Ho Song
Author_Institution :
Dept. of Electron. & Commun. Eng., Hanyang Univ., Seoul, South Korea
Abstract :
Wavefront parallelization speeds up the deblocking filter in multi-core platforms, but it has a limitation in achieving sufficient parallelism owing to data dependency among macroblocks. To overcome this limitation, we propose a novel approach of parallelizing the deblocking filter by exploiting the conditional data dependency affected by the boundary strength 1for each macroblock edge.
Keywords :
data compression; filtering theory; graph theory; multiprocessing systems; video coding; H.264-AVC standard; boundary strength; conditional data dependency; deblocking filter; graph-based parallelization algorithm; macroblock edge; multicore platforms; wavefront parallelization; Finite impulse response filters; Information filtering; Multicore processing; Parallel processing; Video coding; Video sequences;
Conference_Titel :
Consumer Electronics (ICCE), 2013 IEEE International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4673-1361-2
DOI :
10.1109/ICCE.2013.6486920