search

MPEGGroup / CMAF

Official MPEG repository to discuss public issues on CMAF (ISO/IEC 23000-19)
2 stars 0 forks source link

CTA WAVE: Use of a compatibility brand in a file type box or track type box #22

Closed haudiobe closed 1 year ago

haudiobe commented 3 years ago

https://github.com/cta-wave/CMAF-Byte-Stream/issues/21

cconcolato commented 2 years ago

Looking that the CTA-WAVE issue, the request seems to be :

we recommend the definition and use of a child box in the video sample description in order to signal the presence [of 608/708 captions] embedded in video.

This seems to be an ISOBMFF issue first. @krasimirkolarov @dwsinger ?

dwsinger commented 2 years ago

we should look at the SEIs-on-SEIs as I think we can assert in them that the appropriate SEIs for CEAx08 are present, and the SEI-on-SEI could be in the config array

haudiobe commented 2 years ago

Proposal: 1) It is a new feature that would be worthwhile to be discussed on how to add “video sample description in order to signal the presence [of 608/708 captions] embedded in video.”, However, this is only a specific case as rightfully indicated. 2) There is a proposal to SEI-on-SEI: https://github.com/MPEGGroup/CMAF/issues/22#issuecomment-985786067 into the sample entry to indicated the presence and details on contained SEIs. 3) This is a good feature and we propose to add this to the TUC under “SEI presence signaling in the sample entry” 4) Close the issue.

mikedo commented 2 years ago

The presence of 608/708 (in video) is what the 'ccea' compatible brand does. Additionally useful would be to signal the caption tracks, languages, etc and that's what the strings defined in SCTE 128-1 do. See also DASH-IF ATSC IOP v1.1 (although there are syntax bugs so rely on the work in process SCTE 128-1). We should find place to store the latter strings where they could be extracted to populate (e.g. DASH) manifests.

krasimirkolarov commented 2 years ago

MPEG140: Keep this issue for one more cycle, unless there is further feedback

haudiobe commented 1 year ago

The proposal remains to add this to the TUC.

RufaelDev commented 1 year ago

@mikedo is it possible to map this information in these string from SEI messages in the sample entry ? Storing such SEI message in the Sample Entry to describe the SEI messages in the rest of the streams makes a lot of sense to me.

krasimirkolarov commented 1 year ago

We will add that issue to the current CMAF TuC as captured in the BoG notes

krasimirkolarov commented 1 year ago

we want to expose the presence of certain SEIs through ISOBMFF tools which include: