Encapsulating text in RTP stream

[Manuel Wagesreither]

Hi Brad,

Am Do, 7. Jul 2022, um 01:17, schrieb Brad Hards via gstreamer-devel:
> On Wednesday, 6 July 2022 10:37:49 PM AEST Manuel Wagesreither via gstreamer-
> devel wrote:
>> Hi all,
>> 
>> For scientific purposes I'd like to send a video stream together with
>> textual data over an RTP stream. The textual data amend the visual stream
>> and contain information such as X, Y coordinates of points of interest.
>> Thus, they need to be in sync with the video.
>
> You could consider using KLV as metadata stream. That is common in the 
> military context - see https://nsgreg.nga.mil/misb.jsp
>
> If your annotations are in pixel space, consider ST 0602.
>
> If your points of interest are in geographic space, consider ST 0903 or 
> possibly ST 0806.
>
> Note that gstreamer can't demux synchronous metadata, but you could try 
> https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1312

We'll check. However, if demuxing synchronious metadata is not yet supported by gstreamer we'll likely not be able to use this. Our resources are limited and this sounds pretty cutting edge to me.

Regards and thanks for your input,
Manuel