So, Google launched their Image Labeler game the other day. The format is brilliantly simple: players are assigned a random (unknown) partner, with whom they need to collaborate to agree on tags to describe a particular image. It’s a great way to collect metadata about an image - descriptive tags, if you want to put it like that - because you’ve got an ideal situation: a double blind test.

But I come here not to praise Google…

I’ve been noodling on this since I saw it announced - it reminds me of something else I’ve seen. At first I thought it was Fastr, then possibly Peekaboom, but finally I remembered it.

It seems to me to be almost identical to a Carnegie Mellon project which has been running since 2003, called the ESP Game.

From their synopsis:

Labeling an image means associating word descriptions to it, as shown below. Computer programs can’t yet determine the contents of arbitrary images, but the ESP game provides a novel method of labeling them: players get to have fun as they help us determine their contents. If the ESP game is played as much as other popular online games, we estimate that all the images on the Web can be labeled in a matter of weeks!

Having proper labels associated to each image on the Internet would allow for very accurate image search, would improve the accessibility of the Web (by providing word descriptions of all images to visually impaired individuals), and would help users block inappropriate (e.g., pornographic) images from their computers.

Is it a coincidence that Google’s Image Labeller game is so similar? Are they collaborating with Carnegie Mellon? There’s no mention of ESP on the Google game, as far as I can see…