Search Results for "JVET-AD2022"

Found 11 document(s)

Search documents

Use number, keyword, author, or MPEG number. Filter by meeting when needed.

Not signed in yet.
Showing 1-11 of 11 result(s)
30th Meeting: Antalya, TR, April 2023 2023-06-24 13:13
Draft plan for subjective quality testing of FGC SEI message
Authors: Philippe de Lagrange W. Husak Miloš Radosavljević Mathias Wien Jens-Rainer Ohm
Abstract
This document is a draft test plan to assess the visual impact of film grain synthesis, evaluating the potential reduction of bitrate on video content that originally contains grain, maintaining the artistic intent while video coding removes grain, and potentially improving perceived quality by increasing perceived sharpness and hiding coding artefacts in smooth areas. The test planned compares VVC Main 10 profile with and without film grain synthesis, making use of the FGC SEI message.
Meeting Notes

JVET-AD2022 Draft plan for subjective quality testing of FGC SEI message

JVET-AD2022 Draft plan for subjective quality testing of FGC SEI message

ARCHIVE :
JVET-AD2022_v1.docx
Decisions
JVET-AD2022 Draft plan for subjective quality testing of FGC SEI message
JVET-AD2022 Draft plan for subjective quality testing of FGC SEI message
Citation
33rd Meeting: by teleconference, CH, January 2024 2024-01-16 14:37
AHG4/AHG13: source for scanned film test sequences
Authors: Philippe de Lagrange InterDigital
Abstract
This contribution reports about old film scanned by Prelinger Archives and made available on the Internet Archive website, and recommends to select a few samples to be included as candidate test sequences for FGC SEI visual testing activities planned in JVET-AD2022.
Meeting Notes
JVET-AG0228 AHG4/AHG13: source for scanned film test sequences [P. de Lagrange (InterDigital)] [late]

Contribution JVET-AG0228 was discussed at 2115–2135 on Tuesday 23 Jan. 2024 (chaired by JRO and MW) in a joint meeting of JVET and MPEG AG5 Visual quality testing.

This contribution reports about old film scanned by Prelinger Archives and made available on the Internet Archive website, and recommends to select a few samples to be included as candidate test sequences for FGC SEI visual testing activities planned in JVET-AD2022.

It was commented by S. Wenger that a pretty large selection of materials from Internet archive is tagged as public domain, so it could be used. This would however not be the case for all material, would need to be checked on a case by case basis. High quality versions of sequences are also available, though not uncompressed.

It was suggested to take action on selecting some sequences and investigate their compression characteristics. If not possible during the meeting, it should be added to AHG mandates (see AG13 mandates in section 9).

Codec performance with alternative test materials (0)

This section is kept as a template for future use.

AHG5: Conformance test development (1)

Contributions in this area were discussed at 2100–2115 on Thursday 25 Jan. 2024 (chaired by JRO).

JVET-AG0228 AHG4/AHG13: source for scanned film test sequences [P. de Lagrange (InterDigital)] [late]

Contribution JVET-AG0228 was discussed at 2115–2135 on Tuesday 23 Jan. 2024 (chaired by JRO and MW) in a joint meeting of JVET and MPEG AG5 Visual quality testing.

This contribution...

ARCHIVE :
JVET-AG0228-v1.docx
Decisions
This contribution reports about one source where relevant material could be found, and provides information about some samples of this material.
This section is kept as a template for future use.
Citation
33rd Meeting: by teleconference, CH, January 2024 2024-01-25 05:29
[AHG13] Compressibility analysis of Film Grain test sequences
Authors: Seethal Paluri Apple J. Kim Apple Dimitri Podborski Apple Alexis M. Tourapis Apple
Abstract
This document presents results on the compressibility of the film grain content introduced in JVET-AF0262. The sequences are compressed using the FFMPEG software and the encoding statistics are analyzed. Such information could be utilized when selecting content for different experiments or evaluations within JVET or other groups.
Meeting Notes
JVET-AG0290 [AHG13] Compressibility analysis of Film Grain test sequences [S. Paluri, J. Kim, D. Podborski, A. M. Tourapis (Apple)] [late]

This document presents results on the compressibility of the film grain content introduced in JVET-AF0262. The sequences are compressed using the FFMPEG software and the encoding statistics are analyzed. Such information could be utilized when selecting content for different experiments or evaluations within JVET or other groups.

This analysis is deemed to be extremely helpful in the context of selecting test cases for the planned subjective test, to have a variety of content, e.g. with different levels of complexity, and identify which sequences are hard to encode when film grain is present. BoG to prepare version of JVET-AD2022?

In a follow-up discussion, it was suggested to also add some sequences with film grain in CTC, as it is known that they could be difficult to encode (not removing the film grain, not using synthesis).

JVET-AG0290 [AHG13] Compressibility analysis of Film Grain test sequences [S. Paluri, J. Kim, D. Podborski, A. M. Tourapis (Apple)] [late]

This document presents results on the compressibility of the film grain content introduced in JVET-AF0262. The sequences are compressed using the FFMPEG software and the encoding statistics are analyzed. Such information could be utilized when selecting content for different experiments or evaluations within JVET or other groups.

This analysis is deemed to be extremely helpful in the context of selecting test cases for the planned subjective test, to have a variety of content,...

ARCHIVE :
._JVET-AG0290_v1.docx ._JVET-AG0290_v2.docx ._JVET-AG0290_v3.docx ._JVET-AG0290_v4.docx JVET-AG0290_v1.docx JVET-AG0290_v2.docx JVET-AG0290_v3.docx JVET-AG0290_v4.docx
Decisions
This contribution had not been uploaded at the time of preparation of the AHG report.
In a follow-up discussion, it was suggested to also add some sequences with film grain in CTC, as it is known that they could be difficult to encode (not removing the film grain, not using synthesis).
Citation
33rd Meeting: by teleconference, CH, January 2024 2024-01-23 16:56
AHG4/AHG13/AHG3: film grain tuning tool tutorial
Authors: Philippe de Lagrange InterDigital

Abstract not available in document

Meeting Notes
JVET-AG0324 AHG4/AHG13/AHG3: film grain tuning tool tutorial [P. de Lagrange (InterDigital)] [late] This contribution describes how to use the “FGC SEI designer” graphical tool to display and interactively edit an FGC SEI configuration file. That tool is available in the JVET gitlab (in “JVET AHG Film Grain Technologies” group). The intent is to encourage experimentation and showcase the capabilities of the FGC SEI in conjunction with the implementation provided in the same repository, which is asserted to be hardware-friendly and interoperable with different film grain models. It is also asked whether interactive testing based on this tool could be used in the context of AHG4/AG5, to evaluate the grain fidelity aspect considered in the second category of the FGC SEI visual testing plan (JVET-AD2022). Was presented at 2135 on Tuesday Jan. 23 (chaired by JRO and MW)Software tool is availa...
ARCHIVE :
CrowdRun.cfg DinnerScene.cfg Gaussian-1s.cfg JVET-AG0324-v1.zip OldTownCross-2.cfg
Decisions
It was reported that not all film grain on digital materials is created by artists, there are also cameras that insert the film grain right after capture.
Citation
32nd Meeting: Hannover, DE, October 2023 2024-01-17 05:57
Verification test plan for VVC multilayer coding (update 2)
Authors: Shunsuke Iwamura Philippe de Lagrange Mathias Wien Jens-Rainer Ohm
Abstract
This document contains a plan for the video verification test to be conducted to verify the coding performance of the VVC Multilayer Main 10 profile in several configurations: spatial scalability with optimized coding performance, quality ladder involving both spatial and quality scalability, content layering, and stereo coding. The three configurations involve output layer sets with maximum two layers.
Meeting Notes
JVET-AF2021 Verification test plan for VVC multilayer coding (update 2) JVET-AF2021 Verification test plan for VVC multilayer coding (update 2) [S. Iwamura, P. de Lagrange, M. Wien] (2023-12-22) See notes under JVET-AF0311 for updates. Remains valid – not updated: JVET-AD2022 Draft plan for subjective quality testing of FGC SEI message [P. de Lagrange, W. Husak, M. Radosavljević, M. Wien] (2023-06-16) According to discussions under section 4.12, more investigations are necessary on the new sequences. An update of this document was agreed to be postponed.
ARCHIVE :
JVET-AF2021-v1.docx
Decisions
JVET-AF2021 Verification test plan for VVC multilayer coding (update 2)
According to discussions under section 4.12, more investigations are necessary on the new sequences. An update of this document was agreed to be postponed.
Citation
32nd Meeting: Hannover, DE, October 2023 2023-10-18 07:58
BoG on assessment of HOP proposals
Authors: Elena Alshina Franck Galpin
Abstract
This report summarizes the discussion during BoG on assessment of HOP proposals established during JVET-AF meeting.
Meeting Notes
JVET-AF0307 BoG on assessment of HOP proposals [E. Alshina, F. Galpin] This was presented in JVET Wed. 18 Oct. at 1305 (chaired by JRO) The following categories of proposals had been discussed: Filter usage aspects (section 1); Training aspects (section 2); Design elements of HOP for future NNVC (section 3). Recommendations Adoption to NNVC-7.0 Decision (software): Adopt JVET-AF0085 (EE1-1.2.1), enable by default for LOP and HOP; Decision (software fix): Adopt JVET-AF0172 (redundant signalling removal); Decision (software, encoder only): Adopt JVET-AF0193 (disable by default); Decision (training): Adopt JVET-AF0180 and JVET-AF0155 (12:1:1 distortion weight in HOP training) Decision (CTC): Adopt JVET-AF0155 (Chroma QP offset +1 for HOP inference) Next EE1: Agreement: stay with single models for HOP EE1-0: HOP re-train with 12:1:1 loss (in inference Chroma QP offset +1 and JVET-AF0085); JV...
ARCHIVE :
bytedance_data.pptx bytedance_data2.pptx JVET-AF0307-v2.docx tencent_BOG_Suggestions.pptx tencent_Diff_in_EE1-114cd.pptx tencent_Implementation_issue_analysis.pptx
Decisions
adopted
Adopt JVET-AF0205 (decided Thursday 19 Oct. at 1630, after confirmation of successful completion of the crosscheck)
Citation
31st Meeting: Geneva, CH, July 2023 2023-10-06 20:35
Verification test plan for VVC multilayer coding (update 1)
Authors: Shunsuke Iwamura Philippe de Lagrange Mathias Wien Jens-Rainer Ohm
Abstract
This document contains a plan for the video verification test to be conducted to verify the coding performance of the VVC Multilayer Main 10 profile in several configurations: spatial scalability with optimized coding performance, quality ladder involving both spatial and quality scalability, content layering, and stereo coding. The three configurations involve output layer sets with maximum two layers.
Meeting Notes
JVET-AE2021 Verification test plan for VVC multilayer coding (update 1) JVET-AE2021 Verification test plan for VVC multilayer coding (update 1) [S. Iwamura, P. de Lagrange, M. Wien] (2023-09-29) See notes under section 4.5 for updates. Remains valid – not updated: JVET-AD2022 Draft plan for subjective quality testing of FGC SEI message [P. de Lagrange, W. Husak, M. Radosavljević, M. Wien] (2023-06-16) As new sequences were not available yet for viewing during the JVET meeting, an update of this document of this document was agreed to be postponed.
ARCHIVE :
JVET-AE2021-v1.docx
Decisions
JVET-AE2021 Verification test plan for VVC multilayer coding (update 1)
As new sequences were not available yet for viewing during the JVET meeting, an update of this document of this document was agreed to be postponed.
Citation
31st Meeting: Geneva, CH, July 2023 2023-07-17 14:08
Film Grain BOG Report
Authors: W. Husak

Abstract not available in document

Meeting Notes
JVET-AE0294 Film Grain BOG Report [W. Husak (BoG coordinator)] The results of first BoG meeting were presented in JVET on Saturday 15 July at 1445, as shown below. Elements needed included a Denoiser, FG Analyser, and FG Synthesizer. With the GT approach, the denoiser could in the first place be skipped, and it could be investigated whether the artistic film grain comes with the content could be equivalently generated with the FGS SEI. This would not be fully realistic, but denoiser and analysis could be added later. Another application is coding artefact hiding by dithering. In that case, an objective metric is more urgently needed. It was mentioned that in practice also quant matrices are often used together with film grain (example x.265?). The GT data set could be used in the subjective quality testing of SEI for category 2, and also for category 1 -> viewing if possible on Wednesday...
ARCHIVE :
JVET-AE0294.pptx
Decisions
Contributions in this area were discussed at 1630–1640 on Sunday 16 July 2023 and 1400–1410 on Monday 17 July 2023 (chaired by JRO).
noted
The list of JVET ad hoc groups was also issued as a WG 5 output document WG 5 N 235, as noted in section 9.
Citation
30th Meeting: Antalya, TR, April 2023 2023-07-13 10:20
Guidelines for NNVC software development
Authors: Franck Galpin Samuel Eadie Liqiang Wang Zhihuang Xie Ling Li Jens-Rainer Ohm

Abstract not available in document

Meeting Notes
JVET-AD2007 Guidelines for NNVC software development JVET-AD2016 Common test conditions and evaluation procedures for neural network-based video coding technology JVET-AD2017 Common test conditions and evaluation procedures for enhanced compression tool testing JVET-AD2019 Description of algorithms and software in neural network-based video coding (NNVC) version 3 JVET-AD2021 Verification test plan for VVC multilayer coding JVET-AD2022 Draft plan for subjective quality testing of FGC SEI message JVET-AD2023 Exploration experiment on neural network-based video coding (EE1) JVET-AD2024 Exploration experiment on enhanced compression beyond VVC capability (EE2) JVET-AD2025 Algorithm description of Enhanced Compression Model 9 (ECM 9) JVET-AD2027 SEI processing order SEI message in VVC (Draft 4), also issued as WG 5 preliminary WD N 203 JVET-AD2028 Additional conformance bitstreams for VVC mu...
ARCHIVE :
JVET-AD2007-v1.docx
Decisions
”, especially for non-normative, editorial and planning matters
adopted
Adopt JVET-AD0184
Citation
31st Meeting: Geneva, CH, July 2023 2023-07-10 20:48
JVET AHG report: Project Management (AHG1)
Authors: Jens-Rainer Ohm Gary J. Sullivan
Abstract
This document reports on the work of the JVET ad hoc group on Project Management, including an overall status report on JVET’s standardization projects and the progress made during the interim period since the preceding meeting.
Meeting Notes
JVET-AE0001 JVET AHG report: Project Management (AHG1) [J.-R. Ohm (chair), G. J. Sullivan (vice-chair)] Goals and activity The work of the JVET overall had proceeded well in the interim period with lower number of input documents (as compared to the previous meeting) submitted to the current meeting. Intense discussion had been carried out on the group email reflector, and most output documents from the preceding meeting had been produced as planned. Output documents from the preceding meeting had been made initially available at the JVET web site (https://jvet-experts.org/) or the ITU-based JVET site (http://wftp3.itu.int/av-arch/jvet-site/2023_04_AD_Antalya/). It is noted that the previous document sites http://phenix.int-evry.fr/jvet/, http://phenix.int-evry.fr/jct/, and http://phenix.int-evry.fr/jct3v/ are still accessible, but were converted to read-only. The list of documents produ...
ARCHIVE :
JVET-AE0001-v1.docx
Decisions
The AHG recommended its continuation.
Citation
30th Meeting: Antalya, TR, April 2023 2023-04-27 21:14
BoG Report on Metrics for Tool Complexity Assessment
Authors: Xinwei Li Xiang Li
Abstract
This document contains the report of the BoG on metrics for tool complexity assessment. The BoG met on 4/27 Thursday 6:10pm – 7:30pm.
Meeting Notes
JVET-AD0401 BoG Report on Metrics for Tool Complexity Assessment [X. Li] This document contains the report of the BoG on metrics for tool complexity assessment. The BoG met on 4/27 Thursday 6:10pm – 7:30pm. To discuss how to proceed on defining better metrics (In the VVC development, there were other criteria such as number of worst-case memory accesses, number of operations, number of cycles before a certain value would be available, etc) for tool complexity assessment. Recommendations: Increase the precision of runtime ratio by 1 decimal digit in the ECM result template Add a mandate to AHG7 to develop methodology of more reliable runtime measurement Include efficiency vs runtime figures (separated figures for encoding time and decoding time) in AHG7 report Report total memory usage numbers for RA and LDB/P (by the JVET-Z0150 output in encoding log) in AHG7 report and evaluate the vari...
ARCHIVE :
JVET-AD0401-v1.docx
Decisions
No specific discussion was necessary, and this was to be implemented in AHG work. It requires modification of the Excel reporting template.
See notes under JVET-AD0362 for elements included.
Citation
New Search