This is an announcement of WG last call=
 on "RTP Payload Format for Versatile Video Coding (VVC)=E2=80=9D.
 <=
/span>
The document is available for inspection here:<=
/span>
https:=
//datatracker.ietf.org/doc/draft-ietf-avtcore-rtp-vvc/

In response, please state one of the following:

* I support advancing the document to Proposed Sta=
ndard

* I object to advancement to Prop=
osed Standard, due to Issues
described below <Issue description or link>.<=
/o:p>
 =0ABernard Aboba
=0A 
For the=
 Chairs
=
=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=
=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F
jvet mailing list -- jvet@lists=
.rwth-aachen.de
To unsubscr=
ibe send an email to jvet-leave@lists.rwth-aachen.de=

https://lists.rwth-aachen.de/postorius/lists/jvet.lists.rwth-aac=
hen.de
=C2=A0=C2=A0 V=
VC includes support of temporal scalability, by inclusion of the<=
/u>
=C2=A0=C2=A0 signaling of TemporalId in th=
e NAL unit header, the restriction that
=C2=A0=C2=A0 pictures of a particular temp=
oral sublayer cannot be used for inter
=C2=A0=C2=A0 prediction reference by pictu=
res of a lower temporal sublayer, the
=C2=A0=C2=A0 sub-bitstream extraction proc=
ess, and the requirement that each sub-
=C2=A0=C2=A0 bitstream extraction output b=
e a conforming bitstream.=C2=A0 Media-Aware
=C2=A0=C2=A0 Network Elements (MANEs) can =
utilize the TemporalId in the NAL unit
=C2=A0=C2=A0 header for stream adaptation =
purposes based on temporal scalability.
=C2=A0<=
u>
[BA] I'd o=
mit "signaling of" here, since this isn't really signaling. <=
u>
=C2=A0
We can do that, but I will note that, for the high level synta=
x video codec implementers=E2=80=94which realistically are the main audienc=
e for this spec due to the tight coupling between the RTP payload and the h=
igh level syntax implementations=E2=80=94it surely is signaling.<=
/u>
=
=C2=A0
Section 7.2.2.=
1
=C2=A0<=
u>
[BA] This sect=
ion (for temporal) as well as Section 7.2.2.2 (for other scalability)
appears to cov=
er the negotiation of scalability structures. However, no examples of
the negotiatio=
n are provided. This leads me to wonder whether the specification will
in practice re=
quire a profile for seamless SVC interoperability to be achieved.=

=C2=A0
When you talk profile, you mean in the sense of the H.264/SVC =
=E2=80=9Cprofiles=E2=80=9D developed by IMTC/UCIF of yonder?=C2=A0 I wouldn=
=E2=80=99t rule that out.=C2=A0 While we have done what we could to minimiz=
e interop points, we still have way too many to ensure interop of high perf=
ormance systems if they follow different design/implementation strategies f=
or scalability.=C2=A0 I think that=E2=80=99s inevitable given a) the tremen=
dous flexibility of modern video scalability syntax (starting form H.264/SV=
C), and b) the limitations of SDP and offer/answer.=C2=A0 It also hasn=E2=
=80=99t slowed down any industries using scalable video in practice.=C2=A0 =

I will note, though, that an offer-answer negotiation using th=
is payload format and two sensible implementations will always (or almost a=
lways=E2=80=94leaving out corner cases of incompatible video profiles and s=
uch) lead to some interop point, even if that interop would fall back to no=
n-scalable encoding.
=
=C2=A0
=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 the Layer_id in layered coding, but perhaps not in multi=
view
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 coding.=C2=
=A0 (The relevant part of the spec is relatively new,<=
/pre>
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 therefore t=
he soft language).=C2=A0 However, for serious layer pruning,<=
/span>
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 interpretat=
ion of the VPS is required.=C2=A0 We can add language about
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 the need fo=
r stateful interpretation of LayerID vis-a-vis=C2=A0

=C2=A0 =
=C2=A0 =C2=A0 stateless interpretation of TID later."=

=C2=A0

[BA] This looks like an unresolved issue.=C2=A0

=C2=A0
Yes, and this w=
as also commented on by Miska, and I think we have already converged on how=
 to address it.
=C2=A0

Section 7

=C2=A0

There are no complete examples of SDP Offer/Answer n=
egotiation.=C2=A0

=C2=A0
Yes, and in tha=
t we follow what RFC7798 has done.=C2=A0 At the time, it was remarked that =
too many examples can be counter-productive towards a full implementation o=
f the spec.=C2=A0 All too often, implementers take shortcuts
 by copy-pasting code sniplets in specs into their production implementatio=
n and expect that implementation to work.
I=E2=80=99m not=
 fundamentally against adding an example here, but I think 7798 did the rig=
ht thing.=C2=A0 What does the WG think?
=C2=A0=C2=A0 T=
he ideo parameter set (VPS) pertains to a coded video sequences
=C2=A0=C2=A0 (CVS) of multiple layers cove=
ring the same range of access units, and
=C2=A0=C2=A0 includes, among other informa=
tion decoding dependency expressed as
=C2=A0=C2=A0 information for reference pic=
ture list construction of enhancement
=C2=A0=C2=A0 layers.=C2=A0 The VPS provide=
s a "big picture" of a scalable sequence,
=C2=A0=C2=A0 including what types of opera=
tion points are provided, the profile,
=C2=A0=C2=A0 tier, and level of the operat=
ion points, and some other high-level
=C2=A0=C2=A0 properties of the bitstream t=
hat can be used as the basis for session
=C2=A0=C2=A0 negotiation and content selec=
tion, etc.=C2=A0 One VPS may be referenced by
=C2=A0=C2=A0 one or more sequence paramete=
r sets.
=C2=A0<=
u>
How about this=
 instead? 
=C2=A0<=
u>
=C2=A0=C2=A0 &=
quot;The video parameter set (VPS) pertains to a coded video sequence

=C2=A0=C2=A0 (=
CVS) of multiple layers covering the same range of access units, and=

=C2=A0=C2=A0 includes, among other informa=
tion, decoding dependencies expressed as
=C2=A0=C2=A0 information for reference pic=
ture list construction of enhancement
=C2=A0=C2=A0 layers.=C2=A0 The VPS provide=
s a "big picture" of a scalable sequence,
=C2=A0=C2=A0 including what types of opera=
tion points are provided, the profile,
=C2=A0=C2=A0 tier, and level of the operat=
ion points, and some other high-level
=C2=A0=C2=A0 properties of the bitstream t=
hat can be used as the basis for session
=C2=A0=C2=A0 negotiation and content selec=
tion, etc.=C2=A0 One VPS may be referenced by
=C2=A0=C2=A0 one or more sequence paramete=
r sets."
=C2=A0

Th=
is is an announcement of WG last call on "RTP Payload Format for Versa=
tile Video Coding (VVC)=E2=80=9D.
=C2=A0
Th=
e document is available for inspection here:
https://datatracker.ietf.org/doc/draft-ietf-avtcore-rtp-vvc/=

=C2=A0
=C2=A0
WG=
 Last Call will end on August 30, 2021.
=C2=A0
In=
 response, please state one of the following:
=C2=A0
* =
I support advancing the document to Proposed Standard<=
/pre>
=C2=A0
* =
I object to advancement to Proposed Standard, due to Issues
de=
scribed below <Issue description or link>.
=C2=A0
Be=
rnard Aboba
=C2=A0
Fo=
r the Chairs