IIIF AV Demos

[Tom Crane]

Dear all,

The AV working group has collected many examples of the things we need to be able to model in Presentation API 3 – the things found in AV collections. The things we want to open up to the same annotation and interoperability patterns we have for image-based objects.

These use cases led to test manifests and the evolving Presentation 3 spec.

These models need to be tested against reality by verifying that the things we want to do can really be done in a web browser. Annotating content onto a canvas with a time dimension raises technical challenges. [1]

To help explore those challenges, we have been working on a proof of concept with Joscha Jaeger of Filmicweb. This work is funded by the British Library’s Enhancing Discovery and Access for Sound Collections project.

https://github.com/digirati-co-uk/iiif-av-bl

The Github repo includes the test fixtures, implementation details and an evaluation of approaches to media synchronisation and playback.

The proof of concept player can be found here:

https://digirati-co-uk.github.io/iiif-av-bl/

Feedback most welcome!

Tom

 

[1] IIIF AV Technical challenges
https://docs.google.com/document/d/1lcef8tjqfzBqRSmWLkJZ46Pj0pm8nSD11hbCAd7Hqxg/edit

 

[KANZAKI Masahide]

Dear Tom,

Thank you for valuable information. I found some of media urls in fixture manifests are relative to the tool, not manifest itself, hence resolved to unavailable urls (for outside tools).

Could you kindly check those urls ?

Thanks & best regards,

2017年7月28日金曜日 18時05分00秒 UTC+9 Tom Crane:

[Joscha Jaeger]

Dear Masahide,

I changed all paths to fixture manifests and media files into absolute urls.

Best regards,
Joscha

[KANZAKI Masahide]

Hello Joscha, all

With testing against BL's Fixture Manifests (thanks to Joscha), I hopefully managed to make Image Annotator tool to handle those use cases [1].

A few comments that I felt on trying to implement:

- in many cases, textual annotation is put outside the target region (because we want to see what is annotated), and maybe not to be assumed to reside in the box.

- canvas width / height should be provided for video, imho. Dimensions are not required just for rendering, but important if we want to annotate the parts of the video.

- not sure whether the time state should be remembered when switching canvases. Think of changing CDs or DVDs.

- it would be good idea to make sure in A/V spec that end time is "the first time point that is not part of the interval" (i.e. not inclusive) per Media Fragment spec

(note in test 3, synchronised video example, orange and red videos are not located as intended, because the target URIs have double # for the fragments)

cheers,

[1] http://www.kanzaki.com/works/2016/pub/image-annotator?u=/works/2017/test/iiifav-bl-fixture-collection.json

2017年7月31日月曜日 2時14分24秒 UTC+9 Joscha Jaeger: