Difference between revisions of "WNImage"

From CSWiki
Jump to: navigation, search
(Files)
Line 1: Line 1:
 +
== Todo ==
 +
* Rank images for a given synset.
 +
 
== CVS Access ==
 
== CVS Access ==
 
The WNImage tools in the repository under wnimage.
 
The WNImage tools in the repository under wnimage.
Line 11: Line 14:
 
* similarity.py - a library for similarity computations.  Currently just has cosine similarity.
 
* similarity.py - a library for similarity computations.  Currently just has cosine similarity.
 
* captionstools.py - a library for utility functions relating to image caption manipulation (i.e. reading, vector extraction, etc.)
 
* captionstools.py - a library for utility functions relating to image caption manipulation (i.e. reading, vector extraction, etc.)
 +
 +
== Notes ==
 +
JBG has done some basic disambiguation using Lesk (sp?, maximizing definition overlap).

Revision as of 14:43, 4 May 2006

Todo

  • Rank images for a given synset.

CVS Access

The WNImage tools in the repository under wnimage.


Files

  • gen_Xsynsets.py - generates the Xsynset database file for a given list of words on stdin. The db gets output on stdout. The max_depth parameter specifies how many links to follow. It currently only follows hypernyms and it crawls all senses of a word.
  • gen_caption_vects.py - generates a db of the weighted (alpha=0.5) Xsynset for each image caption. pass in the captions file as the first parameter. You need to run this after gen_Xsynsets.py and it expects the result of that to be named Xsynsetdb.py.
  • extract_caption_words.sh - extracts and uniquifies all the words in the captions of a captions file. Input on stdin and output on stdout.
  • Xsynsettools.py - a library for utility functions relating to Xsynsets. Currently just has a function to generate Xsynsets.
  • similarity.py - a library for similarity computations. Currently just has cosine similarity.
  • captionstools.py - a library for utility functions relating to image caption manipulation (i.e. reading, vector extraction, etc.)

Notes

JBG has done some basic disambiguation using Lesk (sp?, maximizing definition overlap).