[IxDA Discuss] A MS Photosynth possibility

[Jeff Howard]

A couple existing computer vision projects come to mind. Like.com
seaches based on visual similarity. The interface allows you to
specify the part of the image that's important for the match. I'm
not sure if this is supported by human curation on the back end or
not.
http://www.like.com/

The closest thing I've seen to uploading your own random image and
finding a match is What the Font, where you upload a typographic
image and it identifies the font for you.
http://www.myfonts.com/WhatTheFont/

// jeff


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://beta.ixda.org/discuss?post=17065