Hi

Jörg 'Yadgar' Bleimann wrote:
> Currently I'm trying to convince a friend of POV-Ray's outstanding
> landscape rendering capabilities by sending him examples from my
> private p.b.i image collection. While browsing through these currently
> about 18,500 images stored on my harddisk since September 2000 when I
> first hit p.b.i, I started to think whether it would be reasonable to
> have all these images stored in a publicly accessible database on
> povray.org... images would be categorized by artist, publication date,
> graphic format, size, total number of pixels, x-y ratio, number of
> colours, color distribution and, mostly important a wide range of
> (possibly hierarchically organized) keywords describing image contents
> and used POV-Ray features and techniques. The images also would be
> logically connected to their scene scripts, as far as available.
> 
> What do you think about such an idea?

First of all, doing something like this is a huge amount of work never 
to be underestimated. I find the idea intriguing and I like to share a 
few ideas of my own regarding this.

Let me first extract some keywords from your post:
* publicly accessible database
* lots of metadata for every image/post
  - extractable from the images: 
    size, ratio, color distribution
  - extractable from the post:
    author, title
  - extractable from scene files:
    used features, POV-Ray version
  - not easily extractable:
    topics, keywords for objects shown in the image
* Connecting images to scene files (linking p.b.i posts to p.b.s-f 
  posts)

I would recommend the following strategy: Intertwine the metadata with 
the semantic web. This gives us a plethora of keywords. For example, 
look at DBPedia. It provides identifiers for everything described by an 
article on Wikipedia. Yes, POV-Ray too: 
http://dbpedia.org/resource/POV-Ray
For Flickr there is already such a database: If you follow this link, 
you will see that there is a photo collection at flickr:
http://www4.wiwiss.fu-berlin.de/flickrwrappr/photos/POV-Ray
I imagine something similar for p.b.i. Every image could be linked with 
appropriate keywords from DBPedia (or other public databases).
Also, they could be linked with descriptions of their authors (name, 
website and so on).

To get back to the points above:
* Integrating the database into the semantic web allows people to make 
arbitrary queries and to combine it with other databases. This is what I 
call "publicly accessible".
* The semantic web has vocabularies for many things: describing images 
(size and so on), indicating the author and title of something, 
capturing metdata of newsgroupsposts, hierarchical keywords for every-
day objects and topics.
* Since the semantic web is all about linking stuff together, linking 
images to newsgroup posts and scene files comes naturally.

Clear drawback: The semantic web is kind of new territory while the 
usual relational database with a web interface is robust and proven.

More about the semantic web: http://www.w3.org/standards/semanticweb/
Especially important is the part about "Linked Data":
http://www.w3.org/standards/semanticweb/data
The integration of flickr metadata into the semantic web is described 
here:
http://wifo5-03.informatik.uni-mannheim.de/flickrwrappr/

Greetings
Urs

Post a reply to this message