Publication detail

H.264 Video Encoder Implementation Impact on Decoded Picture Quality in Error Prone IPTV Environment

SLANINA, M. SHAHID, M.

Original Title

H.264 Video Encoder Implementation Impact on Decoded Picture Quality in Error Prone IPTV Environment

English Title

H.264 Video Encoder Implementation Impact on Decoded Picture Quality in Error Prone IPTV Environment

Type

conference paper

Language

en

Original Abstract

H.264/AVC is currently the most popular standard in use for video encoding. To achieve efficient coding while maintaining the desired level of the visual quality and to combat the distortions that may get introduced while transmission through lossy channels, it is imperative to use the best possible coding methodologies. To this end, it is desirable to ascertain the available implementations of the standard under various conditions of coding and levels of network perturbations. We present here the results of evaluating the behavior of two widely used freely available (JM and x264) and one commercially available (Elecard) H.264/AVC based encoders for encoding of high definition (HD) videos through simulated error-prone networks. For the evaluation, we employed the widely used and well understood peak signal to nois ratio (PSNR). We use the Mann-Whitney test to reveal the differences in objective scores and prove a strong content dependency of results, which prevents us from favoring one implementation above the others. Interestingly, we find that for all implementations, that lossy transmission causes sequences encoded at higher bitrates to give worse scores than low bitrate videos subject to the same distortion.

English abstract

H.264/AVC is currently the most popular standard in use for video encoding. To achieve efficient coding while maintaining the desired level of the visual quality and to combat the distortions that may get introduced while transmission through lossy channels, it is imperative to use the best possible coding methodologies. To this end, it is desirable to ascertain the available implementations of the standard under various conditions of coding and levels of network perturbations. We present here the results of evaluating the behavior of two widely used freely available (JM and x264) and one commercially available (Elecard) H.264/AVC based encoders for encoding of high definition (HD) videos through simulated error-prone networks. For the evaluation, we employed the widely used and well understood peak signal to nois ratio (PSNR). We use the Mann-Whitney test to reveal the differences in objective scores and prove a strong content dependency of results, which prevents us from favoring one implementation above the others. Interestingly, we find that for all implementations, that lossy transmission causes sequences encoded at higher bitrates to give worse scores than low bitrate videos subject to the same distortion.

Keywords

Video encoding, H.264, IPTV, packet loss.

RIV year

2015

Released

06.10.2015

ISBN

978-1-4673-9282-2

Book

2015 7th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)

Pages from

116

Pages to

122

Pages count

7

URL

BibTex


@inproceedings{BUT117303,
  author="Martin {Slanina} and Muhammad {Shahid}",
  title="H.264 Video Encoder Implementation Impact on Decoded Picture Quality in Error Prone IPTV Environment",
  annote="H.264/AVC is currently the most popular standard in use for video encoding. To achieve efficient coding while maintaining the desired level of the visual quality and to combat the distortions that may get introduced while transmission through lossy channels, it is imperative to use the best possible coding methodologies. To this end, it is desirable to ascertain the available implementations of the standard under various conditions of coding and levels of network perturbations. We present here the results of evaluating the behavior of two widely used freely available (JM and x264) and one commercially available (Elecard) H.264/AVC based encoders for encoding of high definition (HD) videos through simulated error-prone networks. For the evaluation, we employed the widely used and well understood peak signal to nois ratio (PSNR). We use the Mann-Whitney test to reveal the differences in objective scores and prove a strong content dependency of results, which prevents us from favoring one implementation above the others. Interestingly, we find that for all implementations, that lossy transmission causes sequences encoded at higher bitrates to give worse scores than low bitrate videos subject to the same distortion.",
  booktitle="2015 7th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)",
  chapter="117303",
  doi="10.1109/ICUMT.2015.7382415",
  howpublished="online",
  year="2015",
  month="october",
  pages="116--122",
  type="conference paper"
}