Re: Video orientation header ext. - what is it supposed to do?

On 27/11/14 13:48, Singh Varun wrote:
> Hi,
>
> The fact it is in the RTP header and the latest version of RTP-usage
> says MUST implement. I suppose the browser should be able to rotate it
> without using CSS rotations?

Sure, the browser would be able to do that. My point is I don't remember 
us discussing *what* the browser should do. And what happens if the app 
also applies an orientation using CSS? There could be a conflict, what 
would take precedence? This must be specced out IMO.

>
>     WebRTC endpoints that send or receive video MUST implement the
>     coordination of video orientation (CVO) RTP header extension as
>     described in Section 4 of [I-D.ietf-rtcweb-video  <https://tools.ietf.org/html/draft-ietf-rtcweb-rtp-usage-21#ref-I-D.ietf-rtcweb-video>].

Thanks, I had missed that.

>
>
>
>
>> On 27 Nov 2014, at 14:37, Stefan Håkansson LK
>> <stefan.lk.hakansson@ericsson.com
>> <mailto:stefan.lk.hakansson@ericsson.com>> wrote:
>>
>> Hi,
>>
>> just noted in [1] that a video orientation RTP header extension seem to
>> be assumed.
>>
>> Has it ever been discussed what that would mean in the application
>> space? I.e. should it transfer to a track property, should it rotate the
>> view shown by the video element automatically (and in that case: how
>> would that interact with CSS rotations), or something else?
>>
>> Sorry if I missed the discussion, but I don't recollect we talked about
>> this.
>>
>> Stefan
>>
>>
>> [1] http://lists.w3.org/Archives/Public/public-webrtc/2014Nov/0078.html
>>
>


Received on Thursday, 27 November 2014 12:54:15 UTC