Comments on: Google Similar Images : How does it work?

By: nour

nour — Sun, 24 May 2009 13:48:45 +0000

“It also shows pretty clearly that GSI is getting clues from words associated with images”

That means they name each and every picture? Isn’t that.. Too much work?

By: Links Dumper #1

Links Dumper #1 — Sat, 25 Apr 2009 14:32:44 +0000

[...] Поиск похожих картинок (или не очень похожих) [...]

By: Twitted by ericrumsey

Twitted by ericrumsey — Thu, 23 Apr 2009 21:33:44 +0000

[...] This post was Twitted by ericrumsey @nullvariable Google Similar Images : How does it work? http://tinyurl.com/dgkytk http://twitter.com/ericrumsey/statuses/1597833166 This ping back was generated with the http://real-url.org free Twitter back pinger! This is an automatic message, nobody red it before so sorry for bugs. [...]

By: Kas Thomas

Kas Thomas — Thu, 23 Apr 2009 21:23:10 +0000

Thank you for this excellent post. You’ve done some valuable investigating here. Did you try submitting an exemplar image without “blackbird” in the name of the image? You should rename the original image to something else (like an MD5 hash) and repeat the experiment.

The Google technology strikes me as unsophisticated. It almost has to be since all of the really good ideas are patented already. ;^)

The seminal work in this area, IMHO, was done by Barnsley and Sloan, of Iterated Systems fame:
http://bit.ly/5jgUD

The Barnsley technique decomposes an image into a set of fractal primitives. The primitives embody invariants that can be compared across other sets of such primitives. By matching set to set, images can be matched. It’s a subtle and remarkable technique, the theoretical basis of which is well described in Ning Lu’s “Fractal Imaging” (out of print but well worth hunting down).

The Iterated Systems technology eventually became part of the Mediabin DAM system, which was acquired by Interwoven.

Prior to the advent of fractal techniques, this type of thing was done by means of neural-net technology (AI), first subjecting images to various kinds of canonicalization and invariant-extraction. The motivations and approaches are well captured in Timothy Masters’ books on AI (again well worth hunting down). I also recommend Parker’s “Algorithms for Image Processing and Computer Vision” (Wiley, 1997).

Whatever Google is doing, it’s not magic. The magical stuff was invented 10 and 20 years ago. :^)

By: Daniel Tunkelang

Daniel Tunkelang — Thu, 23 Apr 2009 19:26:07 +0000

Great reverse engineering / explanation, thanks!