International organisation for standardisation organisation internationale de normalisation

36.3Summary of discussions

36.3.1m33207

Disposition: Accepted. To be included in DuI.

Summary: Clarification and correction on audio channel output position.

36.3.2m33292

Disposition: Partially accepted. Address as one of DVB comments

Summary: Similar to DVB liaison.

36.3.3m32943 SegmentList and SegmentTemplate cardinality in MPEG-DASH

Disposition: Accept Section 4 in DuI

Summary: Editorial fixes. The first two already addressed by 3GPP liaison. The last one is a typo.

36.3.4M33472 Corrigenda: on definition of EventType

Desposition: Recommend to fix it in COR1, assuming we get a NB comments.

Summary: conflict between spec and annex b on allowing or not allowing xml element in event

36.3.5m33293 Corrigenda Issues for MPEG-DASH

Disposition: Accepted the text with additional correction in issue 1 and issue 2 to be included in DuI.

Summary:

• 
  Issue 1: need the clarification on the scope of event id. Accepted.
  
• 
  Issue 2: clarification on how to calculate the duration in two cases. Accepted.
  
• 
  Issue 3: Already addressed.
  

36.3.6m33471 On Extended Profile in 23009-1

Disposition: Partially accepted. Inclusion of DASH events and requirement of aligned events in all representations of an adaptation set would be included after unpdating the text. Essential user defined events are out of scope and should be defined by consortia.

Summary: Defined the “essential” events and proposed some profile constraints.

36.3.7m33112 DASH: signaling the continuity of adaptation sets across periods (discussed here since it was requested in DASH-IF Liaison of AMD1)

Disposition: Noted. We don’t have consensus on the signaling method. Will be further studied at the next meeting.

Summary: 3 methods of signaling 1) using adaptation set id, 2) subset id, needs to be redefine as subset has a different purpose in the spec 3) assetIdentifier, needs to be extended to be used at adaptation set level.

36.3.8m33284 Ad Insertion and Period Continuity

Disposition: Noted. We don’t have consensus on the signaling method. Will be further studied at the next meeting.
Summary: Proposes

• 
  a new attribute or supplemental descriptor to signal period continuity
  
• 
  the constraints for adaptation sets’ association
  
• 
  a definition of period continuity in terms of media playback.
  

36.3.9m33208 Comments on MPEG-DASH Amd.2

Disposition: Partially accepted

• 
  Generalized URL parameters:
  
  • 
    Few examples will be added.
    
  • 
    Fix the xlink table as suggested
    
  • 
    Fix the “should” to shall in SRP parsing
    
• 
  SRD:
  
  • 
    Improve the text as suggested.
    

Summary:

• 
  Generalized URL parameters: a) is the supplementary needed? Yes and it is clear how a legacy client (V1) will resolve the query parameters b) Xlink:href semantics is not clear. Clarify that it points to an URLQueryInfo.
  
• 
  SRD element: a) is allowed only at Adaptation level, but req and subrep. Should it be allowed? No it is not needed. B) a should must be a shall c) improved text.
  

36.3.10m33227 4K live streaming to regular mobile devices - DASH SRD-based

Disposition: Noted.

Summary: show a use-case of the SRD.

36.3.11m33199 On Client Authentication and Content Access Authorization

Disposition: Accepted. To be Included in Amd 2.

Recommendation: Add Essential and Supplement descriptors to MPD and period levels in COR1, if we get a NB request.

Summary: The proposed solution was accepted to be included in Amd2 at the last meeting, but was not included. In this contribution, however EssentialProperty is added to MPD level to avoid repeating the same descriptor for all adaptation sets/all periods.

36.3.12m33260 Carriage of Time Variant Encryption and Signature information in ISOBMFF and its Signaling in DASH

Disposition: Noted. Deferred to FF subgroup on the proposed box first.

Summary: proposed carrying the signature information as a metadata track instead of being signalled in MPD.

36.3.13M33481

Disposition: Noted.

Summary: issue of recognizing texture vs depth track. Association ID/Type may be able to address it.

36.3.14m33285 CE Report on SAND Thomas Stockhammer

Disposition: We will continue this CE.

Summary: report of the CE activities. All the following contributions are summarized in the report.

36.3.15m33166 Evaluation of client assisted cache friendliness Rémi Houdaille, Charline Taibi, Stéphane Gouache

Disposition: Noted

36.3.16m33234 Communication Channel for SAND Imed Bouazizi

Disposition: Noted

36.3.17m33248 CE-SAND: Differentiated services in DASH Xin Wang, Yuqun Fan, Lei Zhou

Disposition: Noted

36.3.18m33286 CE SAND: Metrics Thomas Stockhammer

Disposition: Dropped.

36.3.19m33287 CE SAND: Comments Thomas Stockhammer

Disposition: Dropped.

36.3.20m33370 DASH SANE CE: GBR for VBR Media Doug Young Suh, Yongwoo Cho, Kyungmo Park

Disposition: Noted
Overall disposition: reviewed and selected SAND, PED and interfaces. No PED parameter is identified. No proposed interface is also recognized as necessary. The Metric table does not include DASH Metrics of Part 1. Recommend to update the CE description with new usecases as well as trimmed down table. 3 coordinators, each coordinating one part: Metrics (Emmanuel), SAND (Ali) and PED/interfaces (Mary-Luc). Continue the CE.

36.3.21M33465 Report

Disposition: Noted.

Summary: summarizes three proposals stated below. Recommend to close this CE.

36.3.22m33236 CCB CE: updates for a more robust

Disposition: Noted.

Summary: using cookies to track access token with expiry time

36.3.23m33288 CE CBC: Client Control using Dynamic Services

Disposition: Noted.

Summary: authoring content on fly, so the client doesn’t have access to content to be played in future. The MPD will be updated as the client plays the content. For

36.3.24m33341 CE CBC: Detection and Control of Client Behavior

Disposition: Noted

Summary: Using access record to track what has been played by client. Has a signaling to use the access method.

36.3.25m33283 MPEG-DASH and Quality of Experience

Disposition: A liaison to SG12 Q14/12 on DASH activities related to QoE.

Summary: introducing a book “quality of experience”, “Qualinet white paper on quality of experience”. Recommends sending information about our activities on QE to SG12 Q.14/12 which has an active work item on streaming QoE.

36.3.26m33218 On URI signing in DASH

Disposition: noted.

Summary: describes how URI signing works and possible applications in DASH. URI signing is typically used for validating URL for a short amount of time, since other user agent with same ip can use the same signature to access content.

36.3.27m33271 Adaptation of URL Signing in DASH

Disposition: noted.

Summary: describes how use partial URL signing that can be used,

36.3.28m33111 Systems DASH: signaling tile encapsulation modes

Disposition: Noted.

Summary: 4 cases of HEVC tiling and how they can be signaled in MPD using SRD descriptor. 1) Independent tiles 2) tiles in one file but independent 3) dependent tiles in one file 4) group of tiles (combination of 2 and 3).

36.3.29m33210 Usage of DASH SRD for HEVC Tiling

Disposition: Disposition: Accepted to add SRD descriptor to sub-representation level in AMD2. We don’t need any additional signaling. However, we need a clarification that if one rep is being referenced twice or more by dependency Id and being pointed more than once when the client streams an dependent layer, it needs to receive nad use it one and not multiple times to build the compliant bit-stream. A NB comment on COR1 is welcomed to clarify this point.
Summary: 3 scenarios: 1) independent tiles, 2) HEVC tiled videos, 2a) no dependency between tracks , using depenencyId the dependencies between tracks can be signaled. 2b) tile tracks are dependent to base track. Proposes an adaptation set with header information with not actual image/video and another virtual adaptation set that includes all tiles. 3) Scalable HEVC. Very similar to 2b.
Summary: 2 cases: 1) all tile tracks are the same file 2) ease tile track is in separate file.

Proposed to use sub-representation per tile for case 1 and adaptation sets for case 2. Proposes the SRD to be used in representation and sub-representation.

36.3.30m33211 Dependency type signaling in DASH

Disposition: Noted. Accepted to be included in AMD2.

Summary: propose association id with association type.

36.3.31m33247 Signalling of Timed Metadata in DASH

Disposition. Noted. Accepted as subset of m33211.

Summary: propose association ID.

36.3.32m33216 Mapping of DASH Role to HTML 5

Disposition: Accepted to be included in Amd2.

Summary: mapping role to kind.

36.3.33m33233 HTTP Streaming for DASH

Disposition: Start a CE on HTTP 2.0 and Web sockets, Imed and Kevin S.

Summary: proposes the use of web sockets.

36.3.34m33275 Extending Representation Format for Media Preparation

Disposition: Noted. Needs further exploration.

Summary: suggests to expand URI to other URI than HTTP. Addressing segment by self-addressing metadata.

36.3.35m33281 Top-level Descriptor for MPEG-DASH

Disposition: Accepted and addressed.

Summary: proposing to signal descriptors also at higher level. We already discussed the solution and accepted a solution.

36.3.36m33312 DASH: Mediaprinting for Content Component Copyright Verification

Disposition: Partially accepted. Revised the requirement to : This standard shall support signaling schemes for identifying ownership of content components.

Summary: proposes the new requirement for conveying content copyright verification information. We believe that DASH is for signaling information and not verification or authorization. A revised requirement will be reviewed and added if approved by system.

36.3.37M33172 Scalable HEVC (SHVC) for DASH

Disposition: Noted.

Summary: proposes a special representation to do fast start up and use ARC for the first segment of that representation. It proposes two new attributes for representation to signal. This can be signalled using supplement descriptor instead of new attribute.

36.3.38m33289 Robust DASH-based Live Services

Disposition: Late submission. Will review at next meeting if re-submitted ontime.