JVET-S0004 JVET AHG report: Test material and visual assessment (AHG4) [V. Baroncini, T. Suzuki, M. Wien, A. Norkin, A. Segall, Y. Ye]

Dedicated AHG meeting sessions were held on 2020-05-15 for SDR [JVET-S0041, JVET-S0043], on 2020-05-27 for 360° video [JVET-S0146, JVET-S0149], and on 2020-05-29 for HDR [JVET-S0151, JVET-S0153]. In the SDR category, a close-to-final definition of test sequences and selection of rate points for VTM and HM in the UHD Random Access test case are presented. For the HD Low Delay test case, the identification of test sequences is in progress. For the HDR category, a candidate set of five QP and five HLG sequences with corresponding QP settings are proposed for visual inspection. A tone mapping procedure has also been determined to enable the viewing of the PQ content on available UHD displays. In the 360° video category, viewpaths have been designed. The applicable projection formats for the sequences have to be determined in order to define VTM and HM QPs for the rate points to be assessed visually. A detailed update on the status of the verification test preparation status is provided in JVET-S0253.

The test sequences used for CfP/CTC are available on ftp://jvet@ftp.ient.rwth-aachen.de in directory “/jvet-cfp” (accredited members of JVET may contact the JVET chairs for login information).

Due to copyright restrictions, the JVET database of test sequences is only available to accredited members of JVET (i.e. members of ISO/IEC MPEG and ITU-T VCEG).

It is highlighted that the 12 new HDR-HLG sequences were received during the interim period. These sequences are further described in JVET-S0218.

Structure at test sequence repository

In the last meeting, the directory structure was approved as follows.

• ctc/ Contains the active test set of the common testing conditions
• ahg/ Contains subdirectories with sequences under consideration.
• ce/ Contains subdirectories for data exchange for specific CE
• jvet-cfe/ The sequences used for CfE
• jvet-cfp/ The sequences used for CfP
• old/ Contains the JEM bitstreams directory, used before the CfP
• upload Stays as before
• In the CTC directory, following subdirectories have been created:
• ctc/360/
• ctc/hdr/
• ctc/scc/
• ctc/sdr/

The following related contributions were submitted.

• JVET-S0041 “Status Report on SDR Verification Test Preparation” [M. Wien (RWTH), V. Baroncini (VABTECH ltd)]
• JVET-S0043 “Agenda and Report of the AHG4 Meeting on the SDR Verification Test on 2020-05-15” [M. Wien (RWTH), V. Baroncini (VABTECH), T. Suzuki (Sony)]
• JVET-S0146 “Status Report on 360º video Verification Test Preparation” [M. Wien (RWTH), V. Baroncini (VABTECH ltd), Y. Ye (Alibaba)]
• JVET-S0149 “Agenda and Report of the AHG4 Meeting on the 360 Verification Test on 2020-05-27” [M. Wien (RWTH), Y. Ye (Alibaba), V. Baroncini (VABTECH ltd), T. Suzuki (Sony)]
• JVET-S0151 “Status Report and Proposed Agenda for the AHG4 Meeting on HDR Verification Test Preparation” [A. Segall (Sharp)]
• JVET-S0153 “Agenda and Report of the AHG4 Meeting on the HDR Verification Test on 2020-05-29” [A. Segall, M. Wien, V. Baroncini, T. Suzuki]
• JVET-S0218 “4K HLG test sequences for HDR verification test” [S. Iwamura, S. Nemoto, A. Ichigaya (NHK)]
• JVET-S0246 “Results of dry run subjective assessment of SDR UHD verification test” [V. Baroncini, M. Wien]
• JVET-S0253 “Update on Verification Test Preparation” [M. Wien, V. Baroncini, A. Segall, Y. Ye]

The AHG recommended:

• To review the input contributions related to the verification test preparation at an early stage of the meeting.
• To continue to discuss and to update the verification test plan
• To create a BoG to progress the update of the verification test plan.
• To collect volunteers to conduct the verification test, including volunteers to encode.
• To review the set of available test sequences for the verification tests and potentially collect more test sequences with a variety of content.
• To continue to collect new test sequences available for JVET with appropriate licensing statements.

It was asked what studies could be conducted for determining scalable coding performance (e.g. relative to simulcast and relative to single-layer).

It was commented that verification testing effort has been planned for

• SDR
• HDR
• 360°

Verification testing should also be conducted for all three of the following areas:

• Screen content
• Scalability
• Camera-captured 4:4:4 content