[Sigia-l] Video metadata

[Livia Labate]

> One emerging area that's has been hard for auto-metadata generation and
> parsing is video. DARPA gave us the internet and now the security industry
> may give us auto-metadata for video:

Nice, that's the hard kind of metadata to extract from video. I can see 
how that would be a good complement to the "easy" kind of auto-metadata 
you could generate from speech-to-text (to metadata).

I always use the example of a two-minute video with my mother cooking my 
favorite dish, boeuf a la moutarde. She talks about Dijon but never says 
the word mustard. Could I come across this video if you searched for 
'mustard'?

Yes if there was a controlled vocabulary, but unlikely. If we had the 
technology to extract more contextual references like the 
security/surveillance systems are doing now, we could certainly use this 
information to enrich the existing metadata.