INTERNATIONAL ORGANISATION FOR STANDARDISATION

ORGANISATION INTERNATIONALE DE NORMALISATION

ISO/IEC JTC1/SC29/WG11

CODING OF MOVING PICTURES AND AUDIO

ISO/IEC JTC1/SC29/WG11 N2337

July 1998

Source: Test Subgroup
Purpose: Information
Subject: FAQ The Formal Verification Tests On MPEG-4 Video Error Resilience

 

  1. What is the role of Test Subgroup in MPEG?
    Test Subgroup is in charge of carrying out formal subjective tests on audio and video encoded material to assess the performance of MPEG standards
  2. What is a formal subjective test?
    A formal subjective test is an experiment carried out according to rigorous protocols based on psychophysical methodologies. In these tests subjects are generally asked either to see sequences or listen excerpts and evaluate the quality of the material presented. The test methods used by MPEG are in general described either in ITU-T or ITU-R Recommendations, but when standard methodologies are not appropriate to the goal of the standard, the Test Subgroup develops and validates new methods.
  3. What is the purpose of verification tests?
    In the final phase of the development of its standards (usually when they reach the status of Committee Draft), MPEG wishes to obtain rigorous evaluation of the work done by means of formal subjective tests, called verification tests. In these tests the new standard is usually compared against existing standards.
  4. What are the verification tests planned for MPEG-4?
    A first round of verification tests will be carried out by March ’99, either on the Audio or Video part.
    The planned tests on the video part are:
    • error resilience
    • content-based coding
    • scalability

The planned tests on the audio part are:

5. What are the main outcomes of the narrowband audio broadcasting test?
The goal of this test was to get an impression of the coding efficiency and the coding gain of the MPEG-4 audio tools for broadcasting on AM modulated bands, like HF, MF and LF. Test results showed that MPEG-4 technology can provide significantly better quality with respect to analogue AM transmission. Another important outcome of this test is that AAC non scaleable codec performs better than scaleable MPEG-4 codecs. However scaleable codecs provide higher quality than that achievable in simulcast.