Efficient Parallelization of different HEVC decoding stages

Anand Meher Kotra (1) , Mickael Raulet (1) , Olivier Déforges (1)

1 IETR - Institut d'Électronique et des Technologies du numéRique

In this paper we present efficient parallelization implementations for different stages of the HEVC decoder, which are LCU decoding, deblocking filtering and SAO filtering. Each of the stages are parallelized in separate passes. The LCU decoding is parallelized using Wave front Parallel Processing (WPP). Deblocking and SAO filtering are parallelized by segmenting each picture into separate regions of consecutive LCU rows and processing each of the regions in a concurrent fashion. On a 6 core machine with 6 threads running concurrently, experimental results showed an average accelerating factor of 4.6, 5, 5.35 for the LCU decoding stage and 4.5, 4.9, 5 for deblocking filtering stage and 4, 4.5 and 5 for SAO filtering stages on HD, 1600p and 2160p sequences respectively.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Anand Meher Kotra, Mickael Raulet, Olivier Déforges. Efficient Parallelization of different HEVC decoding stages. DCC, Mar 2013, Snowbird, United States. pp.502-508, ⟨10.1109/DCC.2013.82⟩. ⟨hal-00920909⟩