W3C home > Mailing lists > Public > public-tt@w3.org > January 2013

RE: Re: Re: TTML and aspect ratio

From: John Birch <John.Birch@screensystems.tv>
Date: Thu, 31 Jan 2013 16:00:21 +0000
To: David Ronca <dronca@netflix.com>, "public-tt@w3.org" <public-tt@w3.org>
Message-ID: <0981DC6F684DE44FBE8E2602C456E8ABAAF02D09@SS-IP-EXMB-01.screensystems.tv>
Hi David,

In EBU-TT (a formal open standard subset of TTML with extensions):

Cell and percentage positioning of regions is recommended (extract from Part 2 draft document):
" The EBU-TT format supports three potential mechanisms for positioning text content over an associated video presentation: the 'cell' measurement unit, percentages and the 'pixel' measurement unit.  While these positioning strategies could all be used simultaneously within a single document, the recommended conversion strategy uses the EBU-TT percentage measurement unit to position the regions used for subtitle display. Using percentages is a more universal strategy, since even if the absolute dimensions of the associated video are changed (e.g. by scaling); the relative dimensions are still valid.  To retain information about the assumed presentation, a tts:extent attribute value should be declared in the root element of EBU-TT documents...".

NOTE: In practise, the tts:extent value is for information only when using percentages (or cells).

In EBU-TT Part 1:
" ...TTML defines a region that establishes a coordinate system into which content regions are placed. In EBU-TT the root container region is the active video (see "Definition of terms")."
" ...The term "active video" (known alias: Production Aperture) refers to the portion of the video signal that is used to carry picture information, as specified in SMPTE ST 2016-1:2009 Chapter 4 [10].

NOTE: The active video is distinct from the active image.
"The term "active image" refers to the portion of the video picture area that is being utilized for programme content, as specified in SMPTE ST 2016-1:2009 Chapter 4. The active image excludes letter-box bars and pillar-box bars."

The combination of a root container that is locked to the extent of the active video and using percentages to position regions means that presentation scaling of the video implies scaling of the timed text content.
By adding metadata about the target aspect ratio and the tts:extent of the original video (together with optional information about letterboxing) the possibility also exists for intermediate document processing to re-format the timed text content (in terms of region position, size or text wrapping) **intelligently**.

Best regards,
John Birch

John Birch | Screen | Strategic Partnerships Manager
Main Line : +44 1473 831700 | Ext : 270 | Direct Dial : +44 1473 834532
Mobile : +44 7919 558380 | Fax : +44 1473 830078
John.Birch@screensystems.tv | www.screensystems.tv | https://twitter.com/screensystems

Visit us at
BVE, London Excel, 26-28 February, Stand J02

P Before printing, think about the environment-----Original Message-----
From: David Ronca [mailto:dronca@netflix.com]
Sent: 31 January 2013 14:46
To: public-tt@w3.org
Subject: Re: Re: TTML and aspect ratio

>The implementer of the client provides it. The implementer might ask
>the user, or the system platform might define. Or the TTML client might
>be embedded in a larger context (e.g., an HTML5 browser) that provides
>its own mechanism for defining where to present captions.

 A proprietary model will not work.  We need a method to define a consistent layout for *any* well behaved client.  Consider two scenarios:  1) we implement a client to present the W3C-tt, 2) we pass the W3C-tt to a client that will do the presentation (i.e. game console, etc.).  For both of these scenarios, the client *must* present the captions in a proper and consistent way.  This is our use case, and it does not work if TT presentation is based on an undefined "external authoring context".

We believe that our use case can be accomplished as follows:
1) Cell or percentage positioning
2) default <tt tts:extent>
3) The addition of root container aspect ratio (along with a default behavior for centering the root container in the video window rule).



This message may contain confidential and/or privileged information. If you are not the intended recipient you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. Screen Subtitling Systems Ltd. Registered in England No. 2596832. Registered Office: The Old Rectory, Claydon Church Lane, Claydon, Ipswich, Suffolk, IP6 0EQ
Received on Thursday, 31 January 2013 16:00:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 31 January 2013 16:00:51 GMT