High-Performance Implementation of Stream Model Based H.264 Video Coding on Parallel Processors - Multimedia and Signal Processing - page 427

5 Conclusion

In this paper, we have proposed a set of novel model-based technique called stream-
ing for H.264 encoding, and mapped the stream encoder onto concrete processors.
Our results on several programmable processors show that the stream encoder can
achieve significant performance acceleration. In particular, real-time encoding of
high-definition H.264 video (up to 1080p) is achieved on a typical stream processor
and a desktop GPU. It makes streaming technique more attractive. To achieve per-
formance comparable with ASIC, using programmable processors is wise with respect
to both time and cost, for this type of media processing whose standards are various
and often varied.
Acknowledgments. This work was supported by the National Natural Science Foun-
dation of China, through grants No. 61033008 and No. 60903041.

References

1. Mattson, P.R.: A Programming System for the Imagine Media Processor. PhD thesis, Stan-
ford University (2002)
2. Wu, N., et al.: Streaming HD H.264 encoder on programmable processors. In: Proc. ACM
Multimedia (2009)
3. Stream Processors Inc. SPI Software Documentation (2008),
http://www.streamprocessors.com
4. NVIDIA, GeForce_GTX_200_GPU_Technical_Brief, NVIDIA, Tech. Rep. (2008)
5. Chen, W.-N., Hang, H.-M., et al.: H.264/AVC motion estimation implmentation on Com-
pute Unified Device Architecture (CUDA). In: IEEE International Conference on Multi-
media and Expo., pp. 697-700 (2008)
6. Lin, Y., Li, P., Chang, C., Wu, C., Tsao, Y., Chien, S.: Multi-pass algorithm of motion es-
timation in video encoding for generic GPU. In: Proc. IEEE International Symposium of
Circuits and Systems (2006)
7. Huang, Y.-L., Shen, Y.-C., et al.: Scalable computation for spatially scalable video coding
using NVIDIA CUDA and multi-core CPU. In: Proc. ACM Multimedia, pp. 361-371
(2009)
8. Ahmad, I., He, Y., Liou, M.L.: Video Compression with Parallel Processing. Parallel
Computing 28(7-8), 1039-1078 (2002)
9. Reference software X264-060805,
http://www.videolan.org/developers/x264.html
10. Xun, C., et al.: A framework for stream programming on DSP. In: Proc. 4th International
Conference on Embedded and Multimedia Computing (2009)
11. Cheung, N.-M., Fan, X., et al.: Video Coding On Multi-Core Graphics Processors. IEEE
Signal Processing, 79-89 (March 2010)