Aumento de Desempenho de Codificação Intra H.264 Usando Arquitetura de GPU NVIDIA CUDA®

  • Ronaldo Husemann UFRGS
  • Anderson Giacomolli UFRGS
  • Augusto Lenz UFRGS
  • Diego Schwingel UFRGS
  • Valter Roesler UFRG

Abstract


The recent growth of high definition multimedia applications imposes several restrictions to traditional ISO MPEG-x e ITU-T H.26x video encoders. Considering that, the most recent international standard, H.264 codec, includes several new algorithms in order to increase compression efficiency and application flexibility. Unfortunately this improved technology demands very high performance to support real-time video encoding. Considering this scenario, the current paper proposes the adoption of the GPU NVidia CUDA® technology as development platform to increase performance of a H.264 intra video encoder. Practical results comparing the proposed solution with the original H.264/AVC reference code points to significant performance gains for high definition computing.
Keywords: H.264 encoding, GPU, GPGPU, parallel programming
Published
2011-10-03
How to Cite

Select a Format
HUSEMANN, Ronaldo; GIACOMOLLI, Anderson; LENZ, Augusto; SCHWINGEL, Diego; ROESLER, Valter. Aumento de Desempenho de Codificação Intra H.264 Usando Arquitetura de GPU NVIDIA CUDA®. In: BRAZILIAN SYMPOSIUM ON MULTIMEDIA AND THE WEB (WEBMEDIA), 17. , 2011, Florianópolis. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2011 . p. 51-54.

Most read articles by the same author(s)