<?xmlversion="1.0" encoding="UTF-8"?>version='1.0' encoding='UTF-8'?> <!DOCTYPE rfc> <?rfc compact="yes"?> <?rfc subcompact="yes"?> <?rfc iprnotified="no" ?> <?rfc strict="yes"?> <?rfc symrefs="yes"?> <?rfc toc="yes"?> <?rfc tocdepth="4"?>[ <!ENTITY nbsp " "> <!ENTITY zwsp "​"> <!ENTITY nbhy "‑"> <!ENTITY wj "⁠"> ]> <rfc xmlns:xi="http://www.w3.org/2001/XInclude" category="exp" docName="draft-ietf-avtext-framemarking-16" number="9626" consensus="true" updates="" obsoletes="" ipr="trust200902"submissionType="IETF">submissionType="IETF" symRefs="true" tocInclude="true" tocDepth="4" version="3" xml:lang="en"> <front> <title abbrev="Video Frame Marking">Video Frame Marking RTP Header Extension</title> <seriesInfo name="RFC" value="9626"/> <author fullname="Mo Zanaty" initials="M" surname="Zanaty"> <organization>Cisco Systems</organization> <address> <postal> <street>170 West Tasman Drive</street> <city>San Jose</city> <region>CA</region> <code>95134</code><country>US</country><country>United States of America</country> </postal> <email>mzanaty@cisco.com</email> </address> </author> <author initials="E." surname="Berger" fullname="Espen Berger"> <organization>Cisco Systems</organization> <address> <email>espeberg@cisco.com</email> </address> </author> <author fullname="Suhas Nandakumar" initials="S" surname="Nandakumar"> <organization>Cisco Systems</organization> <address> <postal> <street>170 West Tasman Drive</street> <city>San Jose</city> <region>CA</region> <code>95134</code><country>US</country><country>United States of America</country> </postal> <email>snandaku@cisco.com</email> </address> </author> <dateday="04" month="March"month="August" year="2024"/><area>Applications</area> <keyword>Internet-Draft</keyword><area>WIT</area> <workgroup>avtcore</workgroup> <!-- [rfced] Please insert any keywords (beyond those that appear in the title) for use on https://www.rfc-editor.org/search. --> <keyword>example</keyword> <abstract> <t>This document describes a Video Frame Marking RTP header extension used to convey information about video frames that is critical for error recovery and packet forwarding in RTP middleboxes or network nodes. It is most useful when media isencrypted,encrypted and essential when the middlebox or node has no access to the media decryption keys. It is also useful for codec-agnostic processing of encrypted or unencrypted media, while it also supports extensions for codec-specific information.</t> </abstract> </front> <middle> <sectiontitle="Introduction"anchor="intro"> <name>Introduction</name> <t>Many widely deployed RTP <xreftarget="RFC3550" />target="RFC3550"/> topologies <xreftarget="RFC7667" />target="RFC7667"/> used in modern voice and video conferencing systems include a centralized component that acts as an RTP switch. It receives voice and video streams from each participant, which may be encrypted usingSRTPSecure Real-time Transport Protocol (SRTP) <xreftarget="RFC3711" />,target="RFC3711"/> or extensions that provide participants with private media <xreftarget="RFC8871" />target="RFC8871"/> via end-to-end encryption where the switch has no access to media decryption keys. The goal is to provide a set of streams back to theparticipantsparticipants, which enable them to render the right media content.InFor example, in a simple video configuration,for example,the goal will be that each participant sees and hears just the active speaker. In that case, the goal of the switch is to receive the voice and video streams from each participant, determine the active speaker based on energy in the voice packets, possibly using the client-to-mixer audio level RTP header extension <xreftarget="RFC6464" />,target="RFC6464"/>, and select the corresponding video stream for transmission to participants; see <xreftarget="rtpswitch" />.</t>target="rtpswitch"/>.</t> <t>In this document, an "RTP switch" is used asa common short termshorthand for the terms "switching RTP mixer", "source projecting middlebox", "source forwarding unit/middlebox" and "video switchingMCU"Multipoint Control Unit (MCU)", as discussed in <xreftarget="RFC7667" />.</t>target="RFC7667"/>.</t> <figuretitle="RTP switch" anchor="rtpswitch"><artwork><![CDATA[anchor="rtpswitch"> <name>RTP Switch</name> <artwork><![CDATA[ +---+ +------------+ +---+ | A |<---->| |<---->| B | +---+ | | +---+ | RTP | +---+ | Switch | +---+ | C |<---->| |<---->| D | +---+ +------------+ +---+]]> </artwork></figure>]]></artwork> </figure> <t>In order to properly support the switching of video streams, the RTP switch typically needs some critical information about video frames in order to start and stop forwarding streams.<list style="symbols"></t> <ul> <li> <!--[rfced] Please review whether "e.g." in the following should instead be "i.e.": Original: Because of inter-frame dependencies, it should ideally switch video streams at a point where the first frame from the new speaker can be decoded by recipients without prior frames, e.g. switch on an intra-frame. --> <t>Because of inter-frame dependencies, it should ideally switch video streams at a point where the first frame from the new speaker can be decoded by recipients without prior frames,e.ge.g., switch on an intra-frame.</t> </li> <li> <t>In many cases, the switch may need to drop frames in order to realize congestion control techniques, and it needs to know which frames can be dropped with minimal impact to video quality.</t> </li> <li> <t>For scalable streams with dependent layers, the switch may need to selectively forward specific layers to specific recipients due to recipient bandwidth or decoder limits.</t></list> </t></li> </ul> <t>Furthermore, it is highly desirable to do this in a payload format-agnostic waywhichthat is not specific to each different video codec. Most modern video codecs share common concepts around frame types and other critical information to make this codec-agnostic handling possible.</t> <t>It is also desirable to be able to do this for SRTP without requiring the video switch to decrypt the packets. SRTP will encrypt the RTP payload formatcontents and consequentlycontents; consequently, this data is not usable for the switching function without decryption, which may not even be possible in the case of end-to-end encryption of private media <xreftarget="RFC8871" />.</t>target="RFC8871"/>.</t> <t>By providing meta-information about the RTP streams outside the encrypted media payload, an RTP switch can do codec-agnostic selective forwarding without decrypting the payload. This document specifies the necessary meta-information in an RTP header extension. </t> </section><section title="Key Words for Normative Requirements"><section> <name>Requirements Language</name> <t> The key words"MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY","<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>", "<bcp14>MAY</bcp14>", and"OPTIONAL""<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as described inBCP 14BCP 14 <xreftarget="RFC2119" />target="RFC2119"/> <xreftarget="RFC8174" />target="RFC8174"/> when, and only when, they appear in all capitals, as shown here. </t> </section><section title="Frame<section> <name>Frame Marking RTP HeaderExtension">Extension</name> <t>This specification uses RTP header extensions as defined in <xreftarget="RFC8285" />.target="RFC8285"/>. A subset of meta-information from the video stream is provided as an RTP header extension to allow an RTP switch to do generic selective forwarding of video streams encoded with potentially different video codecs.</t> <t>The Frame Marking RTP header extension is encoded using the one-byte header or two-byte header as described in <xreftarget="RFC8285" />.target="RFC8285"/>. The one-byte header format is used for examples in thismemo.document. The two-byte header format is used when other two-byte header extensions are present in the same RTPpacket,packet since mixing one-byte and two-byte extensions is not possible in the same RTP packet.</t> <t>This extension is only specified for Source (not Redundancy) RTP Streams <xreftarget="RFC7656" />target="RFC7656"/> that carry video payloads. It is not specified for audio payloads, nor is it specified for Redundancy RTP Streams. The (separate) specifications for Redundancy RTP Streams often include provisions for recovering any header extensions that were part of the original source packet. Such provisions can be followed to recover the Frame Marking RTP header extension of the original source packet. Source packet frame markings may be useful when generating Redundancy RTP Streams; for example, the I (Independent Frame) and D (Discardable Frame) bits, defined in <xreftarget="mandatory-scalable" />,target="mandatory-scalable"/>, can be used to generate extra or no redundancy, respectively, and redundancy schemes with source blocks can align source block boundaries with independent frame boundaries as marked by the I bit. </t> <t>A frame, in the context of this specification, is the set of RTP packets with the same RTP timestamp from a specific RTPsynchronization sourceSynchronization Source (SSRC). A frame within a layer is the set of RTP packets with the same RTP timestamp, SSRC, Temporal ID (TID), and Layer ID (LID).</t> <sectiontitle="Longanchor="mandatory-scalable"> <name>Long Extension for ScalableStreams" anchor="mandatory-scalable">Streams</name> <t>The following RTP header extension isRECOMMENDED<bcp14>RECOMMENDED</bcp14> for scalable streams. ItMAY<bcp14>MAY</bcp14> also be used for non-scalable streams, in which case the TID,LIDLID, and TL0PICIDXMUST<bcp14>MUST</bcp14> be 0 or omitted. The ID is assigned per <xreftarget="RFC8285" />, and thetarget="RFC8285"/>. The length is encoded asL=2 which indicatesfollows:</t> <ul> <li>L=2 to indicate 3 octets of data when nothing isomitted, or L=1omitted,</li> <li>L=1 for 2 octets when TL0PICIDX is omitted,or L=0or</li> <li>L=0 for 1 octet when both the LID and TL0PICIDX areomitted.</t> <figure>omitted.</li></ul> <artwork><![CDATA[ 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=2 |S|E|I|D|B| TID | LID | TL0PICIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ or +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=1 |S|E|I|D|B| TID | LID | (TL0PICIDX omitted) +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ or +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=0 |S|E|I|D|B| TID | (LID and TL0PICIDX omitted) +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+]]></artwork></figure>]]></artwork> <t>The following informationareis extracted from the media payload and sent in the Frame Marking RTP header extension.<list style='symbols'> <t>S:</t> <dl newline="true"> <dt>S: Start of Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 in the first packet in a frame within a layer;otherwise MUSTotherwise, <bcp14>MUST</bcp14> be0.</t> <t>E:0.</dd> <dt>E: End of Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 in the last packet in a frame within a layer;otherwise MUSTotherwise, <bcp14>MUST</bcp14> be 0. Note that the RTP header marker bitMAY<bcp14>MAY</bcp14> be used to infer the last packet of the highest enhancementlayer,layer in payload formats with suchsemantics.</t> <t>I:semantics.</dd> <dt>I: Independent Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 for a frame within a layer that can be decoded independent of temporally prior frames,e.g.e.g., intra-frame, VPX keyframe, H.264IDRInstantaneous Decoding Refresh (IDR) <xreftarget="RFC6184" />,target="RFC6184"/>, or H.265IDR/CRA/BLA/RAPIDR / Clean Random Access (CRA) / Broken Link Access (BLA) / Random Access Point (RAP) <xreftarget="RFC7798" />; otherwise MUSTtarget="RFC7798"/>; otherwise, <bcp14>MUST</bcp14> be 0. Note that this bit only signals temporal independence, so it can be 1 in spatial or quality enhancement layers that depend on temporally co-located layers but not temporally priorframes.</t> <t>D:frames.</dd> <dt>D: Discardable Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 for a frame within a layer the sender knows can bediscarded,discarded and still provide a decodable media stream;otherwise MUSTotherwise, <bcp14>MUST</bcp14> be 0.</t> <t>B:</dd> <dt>B: Base Layer Sync (1bit) - Whenbit)</dt><dd>When the TID is not 0, thisMUST<bcp14>MUST</bcp14> be 1 if the sender knows this frame within a layer only depends on the base temporal layer;otherwise MUSTotherwise, <bcp14>MUST</bcp14> be 0. When the TID is 0 or if no scalability is used, thisMUST<bcp14>MUST</bcp14> be0.</t> <t>TID:0. </dd> <dt>TID: Temporal ID (3bits) - Identifiesbits)</dt><dd>Identifies the temporal layer/sub-layer encoded, starting with 0 for the baselayer,layer and increasing with higher temporal fidelity. If no scalability is used, thisMUST<bcp14>MUST</bcp14> be 0. It is implicitly 0 in the short extensionformat.</t> <t>LID:format. </dd> <dt>LID: Layer ID (8bits) - Identifiesbits)</dt><dd>Identifies the spatial and quality layer encoded, starting with 0 for the baselayer,layer and increasing with higher fidelity. If no scalability is used, thisMUST<bcp14>MUST</bcp14> be 0 or omitted to reduce length. When the LID is omitted, TL0PICIDXMUST<bcp14>MUST</bcp14> also be omitted. It is implicitly 0 in the short extension format or when omitted in the long extensionformat.</t> <t>TL0PICIDX:format.</dd> <dt>TL0PICIDX: Temporal Layer 0 Picture Index (8bits) - Whenbits)</dt><dd>When the TID is 0 and the LID is 0, this is a cyclic counter labeling base layer frames. When the TID is not 0 or the LID is not 0,this indicatesthe indication is that a dependency on the given index, such that this frame within this layer depends on the frame with this label in the layer with a TID 0 and LID 0. If no scalability is used, or the cyclic counter is unknown,this MUSTTL0PICIDX <bcp14>MUST</bcp14> be omitted to reduce length. Note that 0 is a valid index value forTL0PICIDX.</t> </list> </t>TL0PICIDX.</dd> </dl> <t>The layer information contained in the TID and LID convey useful aspects of the layer structure that can be utilized in selective forwarding.</t> <t>Without further information about the layer structure, these TID/LID identifiers can only be used for relative priority of layers and implicit dependencies between layers. They convey a layer hierarchy withTID=0TID = 0 andLID=0LID = 0 identifying the base layer. Higher values of TID identify higher temporal layers with higher frame rates. Higher values of LID identify higher spatial and/or quality layers with higher resolutions and/or bitrates. Implicit dependencies between layers assume that a layer with a given TID/LIDMAY<bcp14>MAY</bcp14> depend onlayer(s)a layer or layers with the same or lower TID/LID, butMUST NOTthey <bcp14>MUST NOT</bcp14> depend onlayer(s)a layer or layers with higher TID/LID.</t><t></t> <t> With further information, for example, possible future RTCPSDESsource description (SDES) items that convey full layer structure information, it may be possible to map these TIDs and LIDs to specific absolute frame rates,resolutions andresolutions, bitrates,as well asand explicit dependencies between layers. Such additional layer information may be useful for forwarding decisions in the RTPswitch,switch but is beyond the scope of this memo. The relative layer information is still useful for many selective forwardingdecisionsdecisions, even without such additional layer information. </t> </section> <sectiontitle="Shortanchor="mandatory-non-scalable"> <name>Short Extension forNon-Scalable Streams" anchor="mandatory-non-scalable">Non-scalable Streams</name> <t>The following RTP header extension isRECOMMENDED<bcp14>RECOMMENDED</bcp14> for non-scalable streams. It is identical to the shortest form of the extension for scalable streams, except the last four bits (B and TID) are replaced with zeros. ItMAY<bcp14>MAY</bcp14> also be used for scalable streams if the sender has limited or no information about stream scalability. The ID is assigned per <xreftarget="RFC8285" />, andtarget="RFC8285"/>; the length is encoded asL=0L=0, which indicates 1 octet of data.</t><figure><artwork><![CDATA[ 0 1 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=0 |S|E|I|D|0 0 0 0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+]]></artwork></figure>]]></artwork> <t>The following informationareis extracted from the media payload and sent in the Frame Marking RTP header extension.<list style='symbols'> <t>S:</t> <dl newline="true"> <dt>S: Start of Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 in the first packet in a frame;otherwise MUSTotherwise, <bcp14>MUST</bcp14> be0.</t> <t>E:0.</dd> <dt>E: End of Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 in the last packet in a frame;otherwise MUSTotherwise, <bcp14>MUST</bcp14> be 0.SHOULD<bcp14>SHOULD</bcp14> match the RTP header marker bit in payload formats with such semantics for marking end offrame.</t> <t>I:frame.</dd> <dt>I: Independent Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 for frames that can be decoded independent of temporally prior frames,e.g.e.g., intra-frame, VPX keyframe, H.264 IDR <xreftarget="RFC6184" />,target="RFC6184"/>, or H.265 IDR/CRA/BLA/IRAP <xreftarget="RFC7798" />; otherwise MUSTtarget="RFC7798"/>; otherwise, <bcp14>MUST</bcp14> be 0.</t> <t>D:</dd> <dt>D: Discardable Frame (1bit) - MUSTbit)</dt><dd><bcp14>MUST</bcp14> be 1 for frames the sender knows can bediscarded,discarded and still provide a decodable media stream;otherwise MUSTotherwise, <bcp14>MUST</bcp14> be 0.</t> <t>The</dd> <dt>The remaining (4bits) -bits)</dt><dd>These are reserved/fixed values and not used for non-scalable streams; theyMUST<bcp14>MUST</bcp14> be set to 0 upon transmission and ignored uponreception.</t> </list> </t>reception.</dd> </dl> </section><section title="Layer ID<section> <name>LID Mappings for ScalableStreams">Streams</name> <t> This section maps the specific Layer ID (LID) information contained in specific scalable codecs to the generic LID and TID fields. </t> <t> Note that non-scalable streams have noLayer ID information and thusLID information; thus, they have no mappings. </t><section title="VP9<section> <name>VP9 LIDMapping">Mapping</name> <t> The VP9 <xreftarget="I-D.ietf-payload-vp9" />target="RFC9628"/> Spatial Layer ID (SID, 3 bits) and Temporal Layer ID (TID, 3 bits) in the VP9 payload descriptor are mapped to the generic LID and TID fields in the header extension as shown in the following figure.</t><figure><artwork><![CDATA[ 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=2 |S|E|I|D|B| TID |0|0|0|0|0| SID | TL0PICIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+]]></artwork></figure>]]></artwork> <t> The S bitMUST<bcp14>MUST</bcp14> match the B bit in the VP9 payload descriptor.</t> <t> The E bitMUST<bcp14>MUST</bcp14> match the E bit in the VP9 payload descriptor.</t> <t> The I bitMUST<bcp14>MUST</bcp14> match the inverse of the P bit in the VP9 payload descriptor.</t><t><!--[rfced] Should "field" or some other noun follow "refresh_frame_flags" in this sentence? Or is this referring to the flags (as the verb "are" is plural)? Original: The D bit MUST be 1 if the refresh_frame_flags in the VP9 payload uncompressed header are all 0, otherwise it MUST be 0. --> <t> The D bit <bcp14>MUST</bcp14> be 1 if the refresh_frame_flags in the VP9 payload uncompressed header are all 0; otherwise, it <bcp14>MUST</bcp14> be 0.</t> <t> The B bitMUST<bcp14>MUST</bcp14> be 0 if the TID is 0;otherwise,if the TID is not 0, itMUST<bcp14>MUST</bcp14> match the U bit in the VP9 payload descriptor. Note:Whenwhen using temporally nested scalability structures as recommended in <xreftarget="scalable-structures" />,target="scalable-structures"/>, the B bit and VP9 U bit will always be 1 if the TID is not0,0 since it is always possible to switch up to a higher temporal layer in such nested structures.</t><t><t>The TID,SIDSID, and TL0PICIDXMUST<bcp14>MUST</bcp14> match the correspondingly named fields in the VP9 payload descriptor, with SID aligned in the least significant 3 bits of the 8-bit LID field and zeros in the most significant 5 bits.</t> </section><section title="H265<section> <name>H265 LIDMapping">Mapping</name> <t> The H265 <xreftarget="RFC7798" />target="RFC7798"/> LayerID (6bits)bits), and TID (3 bits) from theNALNetwork Abstraction Layer (NAL) unit header are mapped to the generic LID and TID fields in the header extension as shown in the following figure.</t><figure><artwork><![CDATA[ 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=2 |S|E|I|D|B| TID |0|0| LayerID | TL0PICIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+]]></artwork></figure>]]></artwork> <!--[rfced] [*AD] We see several (similar) sentences like the example below where it might be difficult for the reader to correclty understand what part(s) of the sentence the keyword MUST applies to. We wonder if a rewrite may be helpful to the reader, possibly using a list... Please see the example below (again, other similar instances exist in the document) and let us know if an update like one of the following might work. Original: The D bit MUST be 1 when the NAL unit header NRI field is 0, or an aggregation packet or fragmentation unit encapsulating only NAL units with NRI=0, otherwise it MUST be 0. Perhaps A (the "when" clause applies to both the D bit being set to 1 or NRI=0): When the NAL unit header NRI field is 0, the D bit MUST be either 1 or an aggregation packet or fragmentation unit encapsulating only NAL units with NRI=0. When the NAL unit header NRI field is not set to 0, the D bit MUST be 0. Perhaps B (the "when" clause only applies to the D bit being 0): The D bit MUST be: -1 when the NAL unit header NRI field is 0, -an aggregation packet or fragmentation unit encapsulating only NAL units with NRI=0, or - 0. --> <t>The S and E bitsMUST<bcp14>MUST</bcp14> match the correspondingly named bits in PACI:PHES:TSCI payload structures.</t> <t>The I bitMUST<bcp14>MUST</bcp14> be 1 when the NAL unit type is 16-23 (inclusive) or 32-34 (inclusive), or an aggregation packet or fragmentation unit encapsulating any of thesetypes, otherwisetypes; otherwise, itMUST<bcp14>MUST</bcp14> be 0. These ranges cover intra (IRAP) frames as well as critical parameter sets(VPS, SPS, PPS).</t>(Video Parameter Set (VPS), Sequence Parameter Set (SPS), Picture Parameter Set (PPS)).</t> <t>The D bitMUST<bcp14>MUST</bcp14> be 1 when the NAL unit type is 0, 2, 4, 6, 8, 10, 12, 14,or38, or an aggregation packet or fragmentation unit encapsulating only thesetypes, otherwisetypes; otherwise, itMUST<bcp14>MUST</bcp14> be 0. These ranges cover non-reference frames as well as filler data.</t> <t>The B bitcan notcannot be determined reliably from simple inspection of payloadheaders, and thereforeheaders; therefore, it is determined by implementation-specific means. For example, internal codec interfaces may provide information to set this reliably.</t><t><t>The TID and LayerIDMUST<bcp14>MUST</bcp14> match the correspondingly named fields in the H265 NAL unit header, with LayerID aligned in the least significant 6 bits of the 8-bit LID field and zeros in the most significant 2 bits.</t> </section><section title="H264-SVC<section> <name>H264 Scalable Video Coding (SVC) LIDMapping">Mapping</name> <t> The following shows H264-SVC <xreftarget="RFC6190" />target="RFC6190"/> Layer encoding information (3 bits for spatial/dependency layer, 4 bits for qualitylayerlayer, and 3 bits for temporal layer) mapped to the generic LID and TID fields.</t> <t>The S, E,II, and D bitsMUST<bcp14>MUST</bcp14> match the correspondingly named bits inPACSIPayload Content Scalability Information (PACSI) payload structures.</t> <t>The I bitMUST<bcp14>MUST</bcp14> be 1 when the NAL unit type is 5, 7, 8, 13,or15, or an aggregation packet or fragmentation unit encapsulating any of thesetypes, otherwisetypes; otherwise, itMUST<bcp14>MUST</bcp14> be 0. These ranges cover intra (IDR) frames as well as critical parameter sets (SPS/PPS variants).</t> <t>The D bitMUST<bcp14>MUST</bcp14> be 1 when the NAL unit headerNRINetwork Remote Identification (NRI) field is 0, or an aggregation packet or fragmentation unit encapsulating only NAL units withNRI=0, otherwiseNRI=0; otherwise, itMUST<bcp14>MUST</bcp14> be 0. The NRI=0 condition signals non-reference frames.</t> <t>The B bitcan notcannot be determined reliably from simple inspection of payloadheaders, and thereforeheaders; therefore, it is determined by implementation-specific means. For example, internal codec interfaces may provide information to set this reliably.</t><figure><artwork><![CDATA[ 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=2 |S|E|I|D|B| TID |0| DID | QID | TL0PICIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+]]></artwork></figure>]]></artwork> </section><section title="H264<section> <name>H264 Advanced Video Coding (AVC) LIDMapping">Mapping</name> <t> The following shows the header extension for H264 (AVC) <xreftarget="RFC6184" />target="RFC6184"/> that contains only temporal layer information.</t> <t> The S bitMUST<bcp14>MUST</bcp14> be 1 when the timestamp in the RTP header differs from the timestamp in the prior RTP sequence number from the sameSSRC, otherwiseSSRC; otherwise, itMUST<bcp14>MUST</bcp14> be 0.</t> <t> The E bitMUST<bcp14>MUST</bcp14> match the M bit in the RTP header.</t> <t>The I bitMUST<bcp14>MUST</bcp14> be 1 when the NAL unit type is 5, 7, or 8, or an aggregation packet or fragmentation unit encapsulating any of thesetypes, otherwisetypes; otherwise, itMUST<bcp14>MUST</bcp14> be 0. These ranges cover intra (IDR) frames as well as critical parameter sets (SPS/PPS).</t> <t>The D bitMUST<bcp14>MUST</bcp14> be 1 when the NAL unit header NRI field is 0, or an aggregation packet or fragmentation unit encapsulating only NAL units withNRI=0, otherwiseNRI=0; otherwise, itMUST<bcp14>MUST</bcp14> be 0. The NRI=0 condition signals non-reference frames.</t> <t>The B bitcan notcannot be determined reliably from simple inspection of payloadheaders, and thereforeheaders; therefore, it is determined by implementation-specific means. For example, internal codec interfaces may provide information to set this reliably.</t><figure><artwork><![CDATA[ 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=2 |S|E|I|D|B| TID |0|0|0|0|0|0|0|0| TL0PICIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+]]></artwork></figure>]]></artwork> </section><section title="VP8<section> <name>VP8 LIDMapping">Mapping</name> <t> The following shows the header extension for VP8 <xreftarget="RFC7741" />target="RFC7741"/> that contains only temporal layer information.</t> <t> The S bitMUST<bcp14>MUST</bcp14> match the correspondingly named bit in the VP8 payload descriptor whenPID=0, otherwisePID=0; otherwise, itMUST<bcp14>MUST</bcp14> be 0.</t> <t> The E bitMUST<bcp14>MUST</bcp14> match the M bit in the RTP header. </t> <t> The I bitMUST<bcp14>MUST</bcp14> match the inverse of the P bit in the VP8 payload header.</t> <t> The D bitMUST<bcp14>MUST</bcp14> match the N bit in the VP8 payload descriptor.</t> <!-- [rfced] Please review whether any of the notes in this document should be in the <aside> element. It is defined as "a container for content that is semantically less important or tangential to the content that surrounds it" (https://authors.ietf.org/en/rfcxml-vocabulary#aside). --> <t> The B bitMUST<bcp14>MUST</bcp14> match the Y bit in the VP8 payload descriptor. Note:Whenwhen using temporally nested scalability structures as recommended in <xreftarget="scalable-structures" />,target="scalable-structures"/>, the B bit and VP8 Y bit will always be 1 if the TID is not0,0 since it is always possible to switch up to a higher temporal layer in such nested structures.</t><t><t>The TID and TL0PICIDXMUST<bcp14>MUST</bcp14> match the correspondingly named fields in the VP8 payload descriptor. </t><figure><artwork><![CDATA[ 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID=? | L=2 |S|E|I|D|B| TID |0|0|0|0|0|0|0|0| TL0PICIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+]]></artwork></figure>]]></artwork> </section><section title="Future<section> <name>Future Codec LIDMapping">Mapping</name> <t>The RTP payload format specification for future video codecsSHOULD<bcp14>SHOULD</bcp14> include a section describing the LID mapping and TID mapping for the codec.</t> </section> </section><section title="Signaling Information"><section> <name>Signaling Information</name> <t>The URI for declaring this header extension in an extmap attribute is "urn:ietf:params:rtp-hdrext:framemarking". It does not contain any extension attributes. </t> <t>An example attribute line in SDP:</t><figure><artwork><![CDATA[ a=extmap:3 urn:ietf:params:rtp-hdrext:framemarking]]></artwork></figure>]]></artwork> </section><section title="Usage Considerations"><section> <name>Usage Considerations</name> <!--[rfced] May we update this sentence as follows for the ease of the reader? Note that the introductory "when" phrase mentions a single frame while the recommendation mentions plural frames: please consider if further updates are necessary. Original: When an RTP switch needs to discard a received video frame due to congestion control considerations, it is RECOMMENDED that it preferably drop frames marked with the D (Discardable) bit set, or the highest values of TID and LID, which indicate the highest temporal and spatial/quality enhancement layers, since those typically have fewer dependenices on them than lower layers. Perhaps A: When an RTP switch needs to discard a received video frame due to congestion control considerations, it is RECOMMENDED that it drop: - frames marked with the D (Discardable) bit set, or -frames with the highest values of TID and LID (which indicate the highest temporal and spatial/quality enhancement layers) since those typically have fewer dependencies on them than lower layers. Perhaps B (to upddate the sg/pl switch): When an RTP switch needs to discard received video frames due to congestion control considerations, it is RECOMMENDED that it drop: - frames marked with the D (Discardable) bit set, or -frames with the highest values of TID and LID (which indicate the highest temporal and spatial/quality enhancement layers) since those typically have fewer dependencies on them than lower layers. --> <t>The header extension valuesMUST<bcp14>MUST</bcp14> represent what is already in the RTP payload.</t> <t> When an RTP switch needs to discard a received video frame due to congestion control considerations, it isRECOMMENDED<bcp14>RECOMMENDED</bcp14> that it preferably drop frames marked with the D (Discardable) bit set, or the highest values of TID and LID, which indicate the highest temporal and spatial/quality enhancement layers, since those typically have fewerdependenicesdependencies on them than lower layers.</t><t><!--[rfced] Please clarify what "and forward the same" means in this text. Original: When an RTP switch wants to forward a new video stream to a receiver, it is RECOMMENDED to select the new video stream from the first switching point with the I (Independent) bit set in all spatial layers and forward the same. --> <t> When an RTP switch wants to forward a new video stream to a receiver, it is <bcp14>RECOMMENDED</bcp14> to select the new video stream from the first switching point with the I (Independent) bit set in all spatial layers and forward the same. <!--[rfced] How may we update this text to more easily illustrate the 1:1 mapping between initialism and expansion? Original: ... source to generate a switching point by sending Full Intra Request (RTCP FIR) as defined in [RFC5104]... Perhaps: ... source to generate a switching point by sending RTCP Full Intra Request (FIR) as defined in [RFC5104]... --> An RTP switch can request that a media sourcetogenerate a switching point by sending Full Intra Request (RTCP FIR) as defined in <xreftarget="RFC5104" />,target="RFC5104"/>, for example. </t><section title="Relation<section> <name>Relation to Layer Refresh Request(LRR)">(LRR)</name> <t>Receivers can use the Layer Refresh Request (LRR) <xreftarget="I-D.ietf-avtext-lrr" />target="RFC9627"/> RTCP feedback message to upgrade to a higher layer in scalable encodings. The TID/LID values and formats used in LRR messagesMUST<bcp14>MUST</bcp14> correspond to the same values and formats specified in <xreftarget="mandatory-scalable" />.target="mandatory-scalable"/>. </t><t>Because<!--[rfced] In the following, are "layer" and "refreshes" redundant with what LRR stands for? Please let us know if any updates are necessary. Original: Because frame marking can only be used with temporally-nested streams, temporal-layer LRR refreshes are unnecessary for frame- marked streams. As expanded it would be: Because frame marking can only be used with temporally nested streams, temporal-layer Layer Refresh Request (LRR) refreshes are unnecessary for frame-marked streams. --> <t>Because frame marking can only be used with temporally nested streams, temporal-layer LRR refreshes are unnecessary for frame-marked streams. Other refreshes can be detected based on the I bit being set for the specific spatial layers. </t> </section> <sectiontitle="Scalability Structures"anchor="scalable-structures"> <name>Scalability Structures</name> <t>The LID and TID information is most useful for fixed scalability structures, such as nested hierarchical temporal layering structures, where each temporal layer only references lower temporal layers or the base temporal layer. The LID and TID information is less useful, or even not useful at all, for complex, irregular scalability structures that do not conform to common, fixed patterns of inter-layer dependencies and referencing structures.ThereforeTherefore, it isRECOMMENDED<bcp14>RECOMMENDED</bcp14> to use LID and TID information for RTP switch forwarding decisions only in the case of temporally nested scalability structures, and it isNOT RECOMMENDED<bcp14>NOT RECOMMENDED</bcp14> for other (more complex or irregular) scalability structures.</t> </section> </section> </section><section title="Security Considerations<section> <name>Security and PrivacyConsiderations" >Considerations</name> <t>Inthe Secure Real-Time Transport Protocol (SRTP) <xref"<xref target="RFC3711"/>,format="title"/>" <xref target="RFC3711"/>, RTP header extensions are authenticated and optionally encrypted <xreftarget="RFC9335" />.target="RFC9335"/>. When unencrypted header extensions are used, some metadata is exposed and visible tomiddle boxesmiddleboxes on the network path, while encrypted media data and metadata in encrypted header extensions are not exposed.</t> <t>The primary utility of this specification is for RTP switches to make proper media forwarding decisions. RTP switches are the SRTP peers of endpoints, so they can access encrypted header extensions, but not end-to-end encrypted private media payloads. Othermiddle boxesmiddleboxes on the network path can only access unencrypted headerextensions,extensions since they are not SRTP peers.</t> <t>RTP endpointswhichthat negotiate this extension should considerwhether thiswhether: </t> <ul><li>this video frame marking metadata needs to be exposed to the SRTP peer only, in which case the header extension can be encrypted;or whether other middle boxesor</li> <li>other middleboxes on the network path also need this metadata, for example, to optimize packet drop decisions that minimize media quality impacts, in which case the header extension can be unencrypted, if the endpoint accepts the potential privacy leakage of thismetadata.metadata.</li> </ul> <t> For example, it would be possible to determine keyframes and their frequency in unencrypted header extensions. This information can often be obtained via statistical analysis of encrypted data. For example, keyframes are usually much larger than other frames, so frame size alone can leak this in the absence of any unencrypted metadata. However, unencrypted metadata provides a reliable signal rather than a statistical probability; so endpoints should take that into consideration to balance the privacy leakage risk against the potential benefit of optimized media delivery when deciding whether to negotiate and encrypt this header extension.</t> </section><section title="Acknowledgements"> <t>Many thanks to Bernard Aboba, Jonathan Lennox, Stephan Wenger, Dale Worley, and Magnus Westerlund for their inputs.</t> </section> <section title="IANA Considerations"><section> <name>IANA Considerations</name> <t>This document defines a new extension URItolisted in theRTP"RTP CompactHeaderExtensions sub-registryHeader Extensions" subregistry of theReal-Time"Real-Time Transport Protocol (RTP)ParametersParameters" registry, according to the following data:</t> <t>Extension URI: urn:ietf:params:rtp-hdrext:framemarkinginfo </t> <t>Description: Frame marking information for video streams </t> <t>Contact: mzanaty@cisco.com </t> <t>Reference: RFCXXXX</t> <t>Note9626</t> </section> </middle> <back> <references> <!-- [rfced] Would you like the references to be alphabetized or left in their current order? --> <name>References</name> <references> <name>Normative References</name> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8285.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6184.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6190.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7741.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7798.xml"/> </references> <references> <name>Informative References</name> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7656.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7667.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6464.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3550.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3711.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5104.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8871.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9335.xml"/> <!-- [I-D.ietf-avtext-lrr]; Companion document --> <reference anchor="RFC9627" target="https://www.rfc-editor.org/info/rfc9627"> <front> <title>The Layer Refresh Request (LRR) RTCP Feedback Message</title> <author initials="J." surname="Lennox" fullname="Jonathan Lennox"> <organization>Vidyo, Inc.</organization> </author> <author initials="D." surname="Hong" fullname="Danny Hong"> <organization>Vidyo, Inc.</organization> </author> <author initials="J." surname="Uberti" fullname="Justin Uberti"> <organization>Google, Inc.</organization> </author> <author initials="S." surname="Holmer" fullname="Stefan Holmer"> <organization>Google, Inc.</organization> </author> <author initials="M." surname="Flodman" fullname="Magnus Flodman"> <organization>Google, Inc.</organization> </author> <date month="August" year="2024" /> </front> <seriesInfo name="RFC" value="9627" /> <seriesInfo name="DOI" value="10.17487/RFC9627"/> </reference> <!-- [I-D.ietf-payload-vp9]; Companion document --> <reference anchor="RFC9628" target="https://www.rfc-editor.org/info/rfc9628"> <front> <title>RTP Payload Format for VP9 Video</title> <author initials="J." surname="Uberti" fullname="Justin Uberti"> <organization>Google, Inc.</organization> </author> <author initials="S." surname="Holmer" fullname="Stefan Holmer"> <organization>Google, Inc.</organization> </author> <author initials="M." surname="Flodman" fullname="Magnus Flodman"> <organization>Google, Inc.</organization> </author> <author initials="D." surname="Hong" fullname="Danny Hong"> <organization>Google, Inc.</organization> </author> <author initials="J." surname="Lennox" fullname="Jonathan Lennox"> <organization>8x8, Inc. / Jitsi</organization> </author> <date month="August" year="2024" /> </front> <seriesInfo name="RFC" value="9628"/> <seriesInfo name="DOI" value="10.17487/RFC9628"/> </reference> </references> </references> <section numbered="false"> <name>Acknowledgements</name> <t>Many thanks to <contact fullname="Bernard Aboba"/>, <contact fullname="Jonathan Lennox"/>, <contact fullname="Stephan Wenger"/>, <contact fullname="Dale Worley"/>, and <contact fullname="Magnus Westerlund"/> for their inputs.</t> </section> <!-- [rfced] We had the following questions related to abbreviations used throughout the document. a) Please note that we have expanded these abbreviations as follows on first use. Please let us know any objections. MCU - Multipoint Control Unit (per RFCEditor: please replace7667) SRTP - Secure Real-time Transport Protocol IDR - Instantaneous Decoding Refresh (per RFC 6184) SDES - source description NAL - Network Abstraction Layer CRA - Clean Random Access BLA - Broken Link Access RAP - Random Access Point AVC - Advanced Video Coidng (per RFC 6184) SVC - Scalable Video Coding (per RFCXXXX6190) PACSI - Payload Content Scalability Information NRI - Network Remote Identification VPS - Video Parameter Set SPS - Sequence Parameter Set PPS - Picture Parameter Set b) Please clarify if/how we may expand the following abbreviations: VPX PACI - is this intentionally different from PACSI? c) Should "intra (IDR)" frames instead be "IDR intra-frames"? This formation occurs twice in this document. d) Please note that the following similar abbreviations appear to be differently treated with regard to punctuation: H264 (AVC) H264-SVC We have expanded the abbreviations on first use, but please let us know if/how these should be made uniform with regard to parens and hypheantion. See also our question regarding H264 vs. H.264. e) We note that in Section 3.3.2, "LayerID" is used. Later, in Figure 8, we see "LayerId" (lowercase d). May these be made consistent? If so, which is preferred? Further, could these actually be made "LID" instead (we see TID in both figures in question, which seems similar)? Please review our related cluster-wide AQ prior to responding. --> <!--[rfced] We had thenumberfollowing questions related to terminology used throughout the document. a) Two questions about the header extension: Should this RTP header extension appear using "Video" throughout? We see both of the following forms. Video Frame Marking RTP header extension vs. Frame Marking RTP header extension Secondly, in the Abstract, we see: Original: This document describes a Video Frame Marking RTP header extension used to convey information about video frames that is critical for error recovery and packet forwarding in RTP middleboxes or network nodes. Is the use of the indefinite article "a" intentional ("a Video Frame Marking RTP header extension")? This seems (possibly) contradictory with the capitalization of the proper noun and use in Section 3 (are there more types of Video Frame Marking RTP header extensions?). Please review. --> <!-- [rfced] Please review the "Inclusive Language" portion of the online Style Guide <https://www.rfc-editor.org/styleguide/part2/#inclusive_language> and let us know if any changes are needed. Updates of thisRFC.</t> </section> </middle> <back> <references title="Normative References"> <?rfc include="reference.RFC.2119"?> <?rfc include="reference.RFC.8174"?> <?rfc include="reference.RFC.8285"?> <?rfc include="reference.RFC.6184"?> <?rfc include="reference.RFC.6190"?> <?rfc include="reference.RFC.7741"?> <?rfc include="reference.RFC.7798"?> </references> <references title="Informative References"> <?rfc include="reference.RFC.7656"?> <?rfc include="reference.RFC.7667"?> <?rfc include="reference.RFC.6464"?> <?rfc include="reference.RFC.3550"?> <?rfc include="reference.RFC.3711"?> <?rfc include="reference.RFC.5104"?> <?rfc include="reference.RFC.8871"?> <?rfc include="reference.RFC.9335"?> <?rfc include="reference.I-D.ietf-avtext-lrr"?> <?rfc include="reference.I-D.ietf-payload-vp9"?> </references>nature typically result in more precise language, which is helpful for readers. Note that our script did not flag any words in particular, but this should still be reviewed as a best practice. --> </back> </rfc>