Google is hoping to improve its image search service by labelling and categorizing all the images in its database. They’ve turned this project into a game where searchers compete against each other to match image labels and categories.

Though the game might sound relatively boring, once you give it a go, your competitive streak will have you addicted in no time.

So how does the game work? According to the Google Image Labeler web page:

You’ll be randomly paired with a partner who’s online and using the feature. Over a 90-second period, you and your partner will be shown the same set of images and asked to provide as many labels as possible to describe each image you see. When your label matches your partner’s label, you’ll earn some points and move on to the next image until time runs out. After time expires, you can explore the images you’ve seen and the websites where those images were found. And we’ll show you the points you’ve earned throughout the session.

The Google Image Labeler concept is based on the ESP Game developed by Carnegie Mellon faculty member Luis von Ahn. Using “Crowdsourcing”, a term coined by Wired magazine earlier this year, Google will use a voluntary internet based labor force to complete its image database categorization.

According to Luis von Ahn, Google could label its entire image database via the Google Image Labeler game in about 2 months. The best part for Google is that the process will be cost free.

The challenge with image search has always been relevance. Unlike traditional search where web page and document content provides a starting point for classification and sorting, images provide limited clues for search engine algorithms. So image and video search are reliant on human labeling to ensure relevant results.

Where other image search facilities have used user tagging to help sort the images, the Google Image Labeler concept removes some of the bias that is inherent when the image source is the image labeler.

So if you’ve got a spare 90 seconds to kill and are sick of your PC version of Solitaire, give the Google Image Labeler a go…