7Plenary Discussions, Joint Meetings, BoG Reports, and Summary of Actions Taken
7.1General
Topics for general discussion at the plenary level:
…
7.2Project development
Joint meetings are discussed in this section of this report. Additional notes on the same topics may appear elsewhere in this report. One such joint meeting was held, as recorded below.
Monday 19 Oct. 16:30–1800
It was reported that the parent bodies had agreed that work on HDR extension development for HEVC (which is to be moved into the JCT-VC, effective at the next meeting
)It was reported that the parent bodies had agreed to conduct an and exploration collaboration for study of potential future video coding technology (for whichin a joint exploration teamactivity for study was agreedto be called the Joint Video Exploration Team – JVET), effective at the current meeting; this is outside the defined scope of the JCT-VC – [Notes to be added/clarified] and reported here only for information purposes.
Profiles:
JCTVC-V0039 New High Throughput Profiles for HEVC [A. Tourapis, X. Yang, D. Singer (Apple)] (inter prediction, a variety of bit depths e.g. 10/14 with high-precision transform, 4:4:4, wavefronts/tiles, for which the proponent advocated for these to be allowed to be used together, screen content features, CABAC bypass alignment)
This was sSupported in principle by the parent bodies –; the JCT-VC was asked to work out any necessary details
C.992 / m37069 AVC profiles (progressive, 4:4:4 8b, 4:2:0 10, still picture, predictive lossless mode); this is [out of the scope of JCT-VC] and reported here only for information purposes as a record of joint parent-level consideration
Some interest was expressed in the proposed "Progressive High 10" and "Progressive High 4:4:4 8" profiles;
Requestfurther input was requested.
m36981 Request for additional profiles inside HEVC [G. Barroux, K. Kazui, K. Takeuchi (Fujitsu)]
m37049 Support for scalable RExt profiles [Stephan Wenger]
This proposes combining RExt and+ scalability from 4:0:0 to 4:4:4, with bit depths 8 to 16
Esp. Monochrome 8, 12 & 16, 4:4:4 8, not enabling chroma format scalability
Need JCT-VC contribution
Decision: Adopt 4 profiles (details to be checked)
m37238 Proposal for the Definition of an Ultra-Low-Latency HEVC Profile for Content Production Applications [Marco Mattavelli, Romuald Mosqueron, Junaid Hamad, Wassim Hamidouche, Olivier Deforges, Laurent Zwhalen, Francois Valadoux, Jarno Vanne, Timo D. Hämäläinen, Jacques Hendrickx, Gabriel Fehervari, Kalle Hyvönen]
JCTVC-V0037 On SCC Level Limits [S. Deshpande (Sharp)] (proposes to allow reduced chroma resolution and reduced bit depth to increase DPB picture capacity)
For further study; no action on the basic proposal
SCC tiles & wavefronts combinations? (can't be combined in current draft)
Decision: Allow for 4:4:4 only
Colour-related (no particular concerns were expressed when these topics were briefly touched upon in the joint discussion)
Correction/clarification (not discussed in this joint discussion due to lack of time and degree of detail that seemed needed for review)
JCTVC-V0035 Generalized Constant and Non-Constant Luminance Code Points (may proceed)
JCTVC-V0038 Effective Colour Volume SEI message (not discussed)
SHM codebase mergeing with HM codebase (not discussed in this joint discussion due to lack of time)
7.4List of actions taken affecting the draft HEVC specification
The following is a summary, in the form of a brief list, of the actions taken at the meeting that affect the draft text of the HEVC specification. Both technical and editorial issues are included (although some relatively minor editorial / bug-fix matters are not listed). This list is provided only as a summary – details of specific actions are noted elsewhere in this report and the list provided here may not be complete and correct. The listing of a document number only indicates that the document is related, not that what it proposes was adopted (in whole or in part).
JCTVC-V0031 Various topics identified by the SCC editors
JCTVC-V0042 Allow zero size of palette predictor initializer in PPS, so initialization can be explicitly disabled at the PPS level.
JCTVC-V0043 Restriction for maximum palette predictor size
JCTVC-V0065 Adopt the modified formula for computing PaletteMaxRun including consideration of copy_above_indices_for_final_run_flag.
JCTVC-V0066 Remove both special treatments of IBC as different from inter in the case of CIP (w.r.t. usage for predicting intra regions and TMVP disabling).
JCTVC-V0048 Change the SPS level IBC mode check aspect to the PPS level; Agree to the MV & referencing picture based constraint rather than AMVR based; if the decoder detects the prohibited case, the decoding process will convert so that list 0 uniprediction is used (and the converted motion data is stored).
JCTVC-V0056 In the case where conversion to uniprediction is performed due to 8x8 biprediction, only do that conversion if TwoVersionsOfCurrDecPicFlag is equal to 1; Editors are asked to consider renaming the TwoVersionsOfCurrDecPicFlag and EightbyEightBiPredInUseforCurrPic (or editorial restructuring to avoid defining those flags).
JCTVC-V0050 about sps_max_num_reorder_pics; about sps_max_dec_pic_buffering_minus1 when IBC is used.
JCTVC-V0057 about sps_max_dec_pic_buffering_minus1 when IBC is used.
For 4:4:4 SCC, wavefronts and tiles can be both used at the same time.
Topics that are not related todirectly for SCC but influence affect the specHEVC specification.
JCTVC-V0011 (SHVC problem reports from AHG)
JCTVC-V0036 (correction/clarification of colour description semantics)
JCTVC-V0062 (fix for parsing of bitstream partition nesting SEI message)
JCTVC-V0064 (fixes for semantics of colour remapping information SEI message)
JCTVC-V0035 (generalized matrix coefficient specification based on colour primaries)
JCTVC-V0039 (six additional high throughput profiles)
JCTVC-V0098 (four profiles that combine scalability with format range extensions)
