Time Waster: Google Image Labeler

Google Image Labeler is an addictive game that entices people like you and me to waste time on the Internet going through images that come up in Google’s image search results and labeling them for fun. You are paired with someone else online, given 2 minutes and away you go typing labels into a text box as fast as you can think of them, getting points when you and your partner’s labels match. After a few attempts I got to a 669 ranking for the day. It’s strange how addictive it is considering your points get you absolutely nothing. The genius (MacArthur Genius) who concocted the game is a professor at Carnegie Mellon named Luis von Ahn. He is speaking at SI on Thursday, I found out about the image labeler via an email advert for his lecture. His research focus is human computation and he proposes games to produce this computation. Below is the description he wrote up for his lecture:

Construction of the Empire State Building: 7 million human-hours. The Panama Canal: 20 million human-hours. Estimated number of human-hours spent playing computer solitaire around the world in one year: billions. A problem with today’s computer society? No, an opportunity.

What if this time and energy could be channeled into useful work? What if people could play computer games and accomplish work without even realizing it? What if billions of people collaborated to solve important problems for humanity or generate training data for computers? My work aims at a general paradigm for doing exactly that: utilizing human processing power to solve computational problems in a distributed manner. In particular, I focus on harnessing human time and energy for addressing problems that computers cannot yet solve. Although computers have advanced dramatically in many respects over the last 50 years, they still do not possess the basic conceptual intelligence or perceptual capabilities that most humans take for granted. By leveraging human skills and abilities in a novel way, I want to solve large-scale computational problems and/or collect training data to teach computers many of these human talents. To this end, I treat human brains as processors in a distributed system, each performing a small part of a massive computation. Unlike computer processors, however, humans require an incentive in order to become part of a collective computation. Among other things, I use online games as a means to encourage participation in the process. In this talk, I will describe my work in the area of Human Computation.