Search Results for "JVET-AN0112"

Found 5 document(s)

Search documents

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

40th Meeting: Geneva, CH, October 2025 2025-10-05 13:51
Abstract
This contribution proposes to always code a significant flag with context to indicate if the coefficient level is zero or not for all coefficient levels within all non-zero subblocks of a TSRC block. In order to guarantee all significant flags coded with context and all context coded bin consumption still follows the current TSRC CABAC bin budget, this contribution proposes to reserve as many context bins as the total number of positions of all non-zero subblocks. The pre-defined CABAC bin budget for a TSRC block is updated accordingly by subtracting the number of reserved context bins and then used as a new CABAC bin budget to control the coding of other context coded syntax. In addition, all significant flags within a subblock are coded together in a separate coding pass before coding any other context coded syntax elements for a coefficient level so that there is no interleaving of context coded bins and bypass coded bins. This contribution also proposes that an adaptive Rice parameter may be used to binarize the remaining of the coefficient levels and Rice parameter is signalled at slice level if needed. For CTC application, adaptive Rice is disabled and Rice parameter 0 is used. On top of ECM-18.0, the simulation results are reported as follows.
JVET-AN0112 Non-EE2: Coding of a coefficient level in TSRC [Y. Yu, L. Xu, J. Gan, H. Yu, D. Wang (OPPO)]

This contribution proposes to always code a significant flag with context to indicate if the coefficient level is zero or not for all coefficient levels within all non-zero subblocks of a TSRC block. In order to guarantee all significant flags coded with context and all context coded bin consumption still follows the current TSRC CABAC bin budget, this contribution proposes to reserve as many context bins as the total number of positions of all non-zero subblocks. The pre-defined CABAC bin budget for a TSRC block is updated accordingly by subtracting the number of reserved context bins and then used as a new CABAC bin budget to control the coding of other context coded syntax. In addition, all significant flags within a subblock are coded together in a separate coding pass before coding any other context coded syntax elements for a coefficient level so that there is no interleaving of context coded bins and bypass coded bins. This contribution also proposes that an adaptive Rice parameter may be used to binarize the remaining of the coefficient levels and Rice parameter is signalled at slice level if needed. For CTC application, adaptive Rice is disabled and Rice parameter 0 is used. On top of ECM-18.0, the simulation results are reported as follows.

AI: Natural content: { 0.00%, 0.00%, 0.03%, 100.1% , 99.8%}

Class F: {-0.12%, -0.15%, -0.27%, 99.1%, 98.1%}

Class TGM: {-0.21%, -0.26%, -0.32%, 99.2%, 100.5%}

RA: Natural...

Decisions
It was agreed to investigate this in an EE, as one EE test together with JVET-AN0087. For low QP results, provide test results should be provided with the JVET-AN0086 encoder change turned on and turned off.
Citation
40th Meeting: Geneva, CH, October 2025 2025-10-06 09:16
Abstract
This contribution crosschecks the results of JVET-AN0112 (Non-EE2 Coding of a coefficient level in TSRC). As of the first version of this cross-check report, the partial results of the AI configuration (only classes F-TGM) are available, which match the results provided by the proponents. Note that the run-times in the first version of this crosscheck report are not fully reliable due to cluster fluctuations.
JVET-AN0284 Crosscheck of JVET-AN0112 (Non-EE2: Coding of a coefficient level in TSRC) [M. Abdoli (Xiaomi)] [late]
References:
Decisions
JVET-AN0284 Crosscheck of JVET-AN0112 (Non-EE2: Coding of a coefficient level in TSRC) [M. Abdoli (Xiaomi)] [late]
Citation
40th Meeting: Geneva, CH, October 2025 2025-10-23 10:39

Abstract not available in document

JVET-AN2041 Preliminary Announcement of JVET AHG17 Meeting in Aachen, DE, during 23-27 February 2026 [J.-R. Ohm, M. Wien] (2025-10-27)

Meeting will be hybrid.

Experts requesting for Schengen visa might want to contact J.-R. Ohm and M. Wien as soon as possible.

Future meeting plans, expressions of thanks, a.o.b., and closing of the meeting

The draft of the WG 5 recommendations (see Annex C) was reviewed and approved in JVET at 1030–1100 on Saturday 11 October. In an update discussed on Sunday, the following changes (which had not yet been included in the version that was presented for information in the MPEG WG’s information exchange meeting on Saturday) were approved:

  • The following recommendation was deemed necessary to be added: “WG 5 notes that action appears to not yet have been taken on some previous requests for public availability of ISO/IEC deliverables without cost, and kindly requests that ISO/IEC 23008-5:2017 (Ed. 2), ISO/IEC 23008-5:2017/Amd.1:2017, ISO/IEC 23008-8:2018 (Ed. 2), ISO/IEC 23008-8:2018/Amd1:2019, and ISO/IEC 23091-2:2025 (Ed. 3) be made freely available, considering that equivalent publications are freely available elsewhere from ITU-T as editions of ITU-T H.265.2, ITU-T H.265.1, and ITU-T H.273.” Experts were also asked to inform J.-R. Ohm and G. J. Sullivan if relevant standards were missing from this list, as far as they would be expected to be made publicly available.
  • The responsible person for delivering 23090-3 is to be changed to Y.K. Wang.
  • Some additions are also necessary to the resolutions of thanks, relating to multi-layer testing and...
Decisions
a) JVET documents
The meeting was closed at 1445 CET on 2025-10-12.
Citation
40th Meeting: Geneva, CH, October 2025 2025-10-07 08:49
Abstract
In TSRC, assigning higher context budget priority to syntax elements with greater coding efficiency has been shown to improve performance, as demonstrated in JVET-AM0062 [1]. Following this principle, this contribution proposes to increase the priority of coeff_sign_flag, which also exhibits relatively high coding efficiency.
JVET-AN0087 AHG12: Budget Control for Transform Skip Residual Coding [R. Xu, W. Zhang, F. Yang (Xidian Univ.), B. Li, F. Xing, P. Han, Z. Wang, W. Song (Hisense)] In TSRC, assigning higher context budget priority to syntax elements with greater coding efficiency has been shown to improve performance, as demonstrated in JVET-AM0062. Following this principle, this contribution proposes to increase the priority of coeff_sign_flag, which also exhibits relatively high coding efficiency. The tests in this proposal are conducted on top of one method described in JVET-AM0062 and the bug fix described in JVET-AN0086. Simulation results based on ECM-18.0 are reported as belowClass F: {-0.10%, -0.02%, -0.04%, xx.x%, xx.x%} Class TGM: {-0.23%, -0.28%, -0.32%, xx.x%, xx.x%}\ RA: Class F: {-0.09%, 0.02%, -0.28%, xx.x%, xx.x%} Class TGM: {-0.29%, -0.29%, -0.35%, xx.x%, xx.x%} Low QP AI (2, 7, 12, 17, 6...
Decisions
Results for all sequences (including natural content) and low QP coding results should be provided.
Citation
40th Meeting: Geneva, CH, October 2025 2025-10-03 09:52
Abstract
This document reports the work of the JVET ad hoc group on enhanced compression beyond VVC capability (AHG12) between the 39th meeting, Daejeon, KR, 26 June - 4 July 2025, and the 40th meeting, Geneva, CH, 3-12 October 2025.
JVET-AN0012 JVET AHG report: Enhanced compression beyond VVC capability (AHG12) [M. Karczewicz, Y. Ye, L. Zhang (co-chairs), B. Bross, R. Chernyak, X. Li, K. Naser, Y. Yu (vice-chairs)] Activities The primary activity of the AHG was the “Exploration experiment on enhanced compression beyond VVC capability” (JVET-AM2024). The combined improvements of the ECM-18.0 over VTM-11.0ecm18.0 anchor for AI, RA and LB configurations are: All Intra Main 10 Y U V EncT DecT Class A1 -14.82% -16.49% -27.44% 1202.2% 577.9% Class A2 -21.47% -24.85% -29.07% 1119.5% 624.0% Class B -15.06% -23.15% -21.18% 1086.0% 627.5% Class C -15.19% -12.08% -13.23% 1026.3% 583.6% Class E -19.58% -23.35% -21.51% 992.5% 615.0% Overall -16.87% -19.90% -21.83% 1080.0% 606.4% Class D -12.92% -8.80% -9.64% 1013.4% 663.9% Class F -30.84% -34.56% -34.60% 688.3% 658.9% Class TGM -43.72% -49.31% -48.60% 513.1% 658.7% Random Access...
Decisions
To review all the related contributions.
Citation
New Search