Visual search, auto-tagging images

On the heels of Riya’s new visual search engine comes another photo-recognition: ALIPRwhich stands for Automatic Linguistic Indexing of Picture – Real time.

It’s interesting to me because visual search is getting a lot of attention right now. But don’t run out and expect it to actually work in real-world situations. ALIPR has a vocabulary of only 332 words right now, almost all simple nouns.

And a real-world test reports mixed results:

In its first real-world test, the program processed thousands of publicly accessible images available on the photo-sharing site Flickr. At least one accurate tag was generated for 98 percent of all the pictures analysed.

“At least one accurate tag” is not a great success rate until we also know how many incorrect tags the software generated … and how the images were selected.

Still, it shows promise:

For 51 percent of these images, the first word generated by ALIPR appeared in users’ tags.

[tags] visual, search, tags, tagging, alipr, riya, flickr, john koetsier [/tags]