Deutsche Telekom AG, T-Labs (Research & Innovation) - ITU-T P.1201.2 audiovisual quality estimation tool

Download Download

This tool is an implementation of the audiovisual coding quality module of the ITU-T P.1201.2 standard. It estimates the audiovisual, video, audio coding quality for IP-based video streaming applications. It works on audiovisual bitstream (PCAP) in combination with the modified JM H.264/AVC codec. It also works with encrypted data, but the I/P/B frame sizes have in that case to be estimated, and the performance of the model depends on the accurate estimation of the I/P/B frame sizes. Finally, video files encoded with x264 or the H.264/AVC JM Reference Software are also supported.

Licences:

For intellectual-property-related information, please refer to <www.itu.int/net4/ipr/>
All rights are reserved.

                                                                                                                                                                 

Provided for research purpose only. For commercial use, please contact Dr. B. Feiten <Bernhard.Feiten@telekom.de>

References:

If you use the tool and publish results, please cite the following papers:

[1] M.-N. Garcia, P. List, S. Argyropoulos, D. Lindegren, M. Pettersson, B. Feiten, J. Gustafsson, A. Raake: “Parametric model for audiovisual quality assessment in IPTVITU-T Rec. P.1201.2”, in Proc. IEEE Int. Workshop on Multi-media Signal Processing (IEEE MMSP), Pula, Italy, 2013
[2] ITU-T Recommendation P.1201.2 (2012)Parametric non-intrusive assessment of audiovisual media streaming quality - Higher resolution application area

Usage example:

./call_P1201_2.py input results.txt

input is a directory containing two types of files (.vfi, .pre) per audiovisual sequence:

  • .pre side information, e.g. video codec and profiles, audio codec, etc. 
  • .vfi list of frame types and sizes

results.txt contains resulting estimated audio, video and audiovisual quality scores per files listed in input

P.1201.2 scores:

On VQEG JEG Hybrid no loss database < ftp://ftp.ivc.polytech.univ-nantes.fr/VQEG/JEG/HYBRID/database>: <ftp://ftp.ivc.polytech.univ-nantes.fr/VQEG/JEG/HYBRID/database/MEASURES/P1201.2>

 

Required software:

python 2.7 and later versions

Reporting a bug:

To report a suspected bug, please contact: marie-neige.garcia@tu-berlin.de

 

Download Download