Re: [AVTCORE] [payload] RFC 6184 H.264 RPSI/SLI usage

"Coban, Muhammed" <> Wed, 20 January 2016 21:54 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id B0CEB1AD0D1; Wed, 20 Jan 2016 13:54:10 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -7.001
X-Spam-Status: No, score=-7.001 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id FkW5NIdO9sQr; Wed, 20 Jan 2016 13:54:08 -0800 (PST)
Received: from ( []) (using TLSv1.2 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 910ED1AD0CC; Wed, 20 Jan 2016 13:54:08 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple;;; q=dns/txt; s=qcdkim; t=1453326848; x=1484862848; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=Ny1q66ezAcE0MjEtle0adEoOiI5GP7btq1tgHc+r/iM=; b=WfTgw5sIOFTJ0d0sB4faWDe6M6OOXvzGz3mB+JL27iWCnPpy3kdXBR46 uzaLkhSWDXf7/t/Mi3bt3xiouQHq+EMtdhiseq9ylRWsv9QXKAqRSkvLE BM0ZHIavIBu86UZip20379ags7LnuA3Z3zwwRGQZonlyKYLNhOUNzWTnl 8=;
X-IronPort-AV: E=Sophos;i="5.22,322,1449561600"; d="scan'208,217";a="107024491"
Received: from ([]) by with ESMTP/TLS/DHE-RSA-AES256-SHA; 20 Jan 2016 13:54:08 -0800
X-IronPort-AV: E=McAfee;i="5700,7163,8050"; a="17920625"
Received: from ([]) by with ESMTP/TLS/RC4-SHA; 20 Jan 2016 13:54:06 -0800
Received: from ( by ( with Microsoft SMTP Server (TLS) id 15.0.1130.7; Wed, 20 Jan 2016 13:54:06 -0800
Received: from ([]) by ([]) with mapi id 15.00.1130.005; Wed, 20 Jan 2016 13:54:06 -0800
From: "Coban, Muhammed" <>
To: Stephan Wenger <>, "" <>, "" <>
Thread-Topic: [payload] [AVTCORE] RFC 6184 H.264 RPSI/SLI usage
Thread-Index: AQHRUyzGMmVE2+Bf9UKWwf+8vpiUb58E6I7A
Date: Wed, 20 Jan 2016 21:54:05 +0000
Message-ID: <>
References: <> <>
In-Reply-To: <>
Accept-Language: en-US
Content-Language: en-US
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: []
Content-Type: multipart/alternative; boundary="_000_216dc03fa19c4996ba8e0703ebe4f14analasexr02enaqualcommco_"
MIME-Version: 1.0
Archived-At: <>
Subject: Re: [AVTCORE] [payload] RFC 6184 H.264 RPSI/SLI usage
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: Audio/Video Transport Core Maintenance <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Wed, 20 Jan 2016 21:54:10 -0000

Thanks Stephan for the clarification. We are interested in defining a clear spec. for interoperability. We have product interests in usage of these messages. Defining frame_num or POC for the codec specific field to be used with RPSI/SLI messages in a paragraph that would update RFC 6184 would be sufficient.


From: Stephan Wenger []
Sent: Tuesday, January 19, 2016 6:46 PM
To: Coban, Muhammed;;
Subject: Re: [payload] [AVTCORE] RFC 6184 H.264 RPSI/SLI usage

Hi Muhammed,
The (codec-) native RPSI bit string of 4585 was meant to be exactly that—a codec-defined bit string.  H.263+ Annex N defines such a bit string in sections N.3.1 and N.4.  There is no equivalent for H.264, so conceptually RPSI cannot be used for H.264.  (Nor for H.265, VPx, …).  If people were interested in reference picture selection based on feedback, they would either create a mapping in the responsible video coding standard group, or use the H.271 route.
As you correctly remark, H.271 includes an RPSI-like equivalent for H.264. I know of one system (and there may have been more) that used the H.271 feedback message for H.264 over H.323/H.245.  I’m not aware of anyone using it over RTCP.  There was an effort underway to create an update for H.271 to support H.265; that effort died the death of lack of resources, but it could be restarted.  ITU-T Q1/16 would be the right group to do so.
As for SLI’s PictureID and H.264, again, there is no defined mapping.  frame_num would probably work in most cases, but to satisfy interop and standardization requirements, that would need to be documented (probably in an RFC updating 6184).
Do you guys have more than academic interest in any of these?  It’s not hard to define them, but would we spend our efforts for anything beyond creation of additional pages of standards (and patents :-)?

From: payload <<>> on behalf of "Coban, Muhammed" <<>>
Date: Monday, January 18, 2016 at 12:28
To: "<>" <<>>, "<>" <<>>
Subject: [payload] [AVTCORE] RFC 6184 H.264 RPSI/SLI usage


There seems to be no specification of Native RPSI bit string and PictureId field definitions for H.264  regarding RFC 4585 RPSI and SLI messages. RFC 6184 doesn’t specify anything for these fields for H.264. Only thing I can find is the RFC 5104 referencing H.271 specification that defines a message that is similar to RPSI for H.264, but I am not sure whether this is widely accepted/used.  What is used for H.264 specific fields for RPSI and SLI?

This issue was raised before, but had no replies on AVTCORE reflector.

Best regards,

Muhammed Coban