“Parallelization of the x264 encoder using OpenCL” by Marth and Marcus
Conference:
Type:
Entry Number: 72
Title:
- Parallelization of the x264 encoder using OpenCL
Presenter(s)/Author(s):
Abstract:
With the introduction of H.264, the complexity on video encoders has increased dramatically. As hardware based encoding solutions profit from the strict sequential design and already feature real time capabilities for high definition material, software solutions lack most of the encoding performance. More precisely, the performance of software encoders is limited due to the computation power of encoding system as well as the high level of codec-intern dependencies. As a consequence, software encoders supporting high definition needs are very rare.
References:
Chen, W.-N., and Hang, H.-M. 2008. H.264/avc motion estimation implementation on compute unified device architecture (cuda). Tech. rep., National Chiao-Tung University.Google Scholar
Cross, J., 2008. GPU Accelerated Video Transcoding. Online Article, December. {online} http://www.extremetech.com/article2/0, 2845, 2337057, 00.asp.Google Scholar
Schwalb, M., Ewerth, R., and Freisleben, B. 2009. Fast motion estimation on graphics hardware for h.264 video encoding. Trans. Multi. 11, 1, 1–10. Google ScholarDigital Library
Shimpi, A. A., 2008. Badaboom: A Full Test of Elemental’s GPU Accelerated H.264 Transcoder. Online Article, August. {online} http://www.anandtech.com/show/2586.Google Scholar
