DocumentCode :
248118
Title :
Collaborative inter-prediction on CPU+GPU systems
Author :
Momcilovic, Svetislav ; Ilic, Aleksandar ; Roma, Nuno ; Sousa, Leonel
Author_Institution :
INESC-ID / IST, Univ. de Lisboa, Lisbon, Portugal
fYear :
2014
fDate :
27-30 Oct. 2014
Firstpage :
1228
Lastpage :
1232
Abstract :
In this paper we propose an efficient method for collaborative H.264/AVC inter-prediction in heterogeneous CPU+GPU systems. In order to minimize the overall encoding time, the proposed method provides stable and balanced load distribution of the most computationally demanding video encoding modules, by relying on accurate and dynamically built functional performance models. In an extensive RD analysis, an efficient temporary dependent prediction of the search area center is proposed, which allows dependency-aware workload partitioning and efficient GPU parallelization, while preserving high compression efficiency. The proposed method also introduces efficient communication-aware techniques, which maximize data reusing, and decrease the overhead of expensive data transfers in collaborative video encoding. The experimental results show that the proposed method is able of achieving real-time video encoding for very demanding video coding parameters, i.e. full HD video format, 64×64 pixels search area and the exhaustive motion estimation.
Keywords :
data compression; graphics processing units; motion estimation; video coding; AVC inter-prediction; CPU-plus-GPU systems; GPU parallelization; H.264 inter-prediction; RD analysis; advanced video coding; central processing unit; collaborative inter-prediction; communication-aware techniques; compression efficiency; dependency-aware workload partitioning; graphics processing unit; motion estimation; search area center; video coding parameters; video encoding modules; Collaboration; Encoding; Graphics processing units; Radio frequency; Real-time systems; Streaming media; Video coding; CPU+GPU computing; divisible load theory; load-balancing; video coding;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing (ICIP), 2014 IEEE International Conference on
Conference_Location :
Paris
Type :
conf
DOI :
10.1109/ICIP.2014.7025245
Filename :
7025245
Link To Document :
بازگشت