that is cool chip. real cool. --- Chip Bloch <wbloch@xxxxxxxxx> wrote: > I received this article on another list that I > belong to. I hope some of you will find it as > interesting as I did. > Chip > > Gamers help the blind get the picture > > * 13:37 16 May 2006 > * NewScientist.com news service > * Paul Marks > > Players are given a description and must then scour > the web for the > correct picture > > Gamers now have the perfect excuse to sit in front > of their computers > all day * they can perform a public service. > > Researchers at Carnegie Mellon University (CMU) in > Pittsburgh, > Pennsylvania, have designed an online game that aims > to harness players' > brainpower to help make websites more accessible to > blind people. > > Visually impaired people often use text-to-speech > converters called > screen readers to listen to the content of web pages > spoken by a > synthesised voice. However, the pictures on most > websites remain > inaccessible because very few have detailed captions > to accurately > describe them. > > The online game "Phetch", which will be made > available at > http://www.peekaboom.org/phetch/, is designed to > encourage other web > users to generate these missing captions. Played in > groups of three to > five people, it randomly assigns the role of > "describer" to one player; > the rest become "seekers". > > > Seek and find > > The game then serves up a randomly chosen website > image to the > describer, who has to write a pithy short paragraph > about it. The words > are then sent to the seekers, who use search engines > to hunt down the > correct picture on the web. The first seeker to find > the image becomes > the describer in the next round. > > If the describer's description is good enough to > lead the seekers to the > picture, it is stored as a caption for that image. > If not, the attempt > is discarded. > > "We hope to collect captions for every image on the > web," says Shiry > Ginosar, a member of the Phetch team. In tests, 130 > players generated > 1400 captions over the course of a week. At this > rate, she says, just > 5000 people could annotate all the pictures indexed > by Google Images in > just 10 months. > > > Web designers > > But Ginosar admits getting web designers the world > over to use the > better captions may be tricky. "We are just > concerned about gathering > caption data right now," she says. > > Julie Howell from the UK's Royal National Institute > for the Blind says > the game addresses a pressing issue. "The web is a > great resource but as > it becomes more picture-led and graphical it should > not become less > accessible for the blind," she told *New Scientist*. > "It's true that > many pictures are simply uncaptioned or just have a > filename." > > The CMU team previously developed another game > "Peekaboom" to help > improve image recognition algorithms. This game > involves two players: > the first must reveal key parts of an image to the > second person, who > must try to guess what is being revealed. The theory > is that players > will reveal the most important parts of an image > first. This could help > computers better identify unfamiliar images by > focusing > > > Blog > > This is a neat idea: > > Images on the Web present a major accessibility > issue for the > visually impaired, mainly because the majority > of them do not have > proper captions. This paper addresses the > problem of attaching > proper explanatory text descriptions to > arbitrary images on the Web. > To this end, we introduce Phetch, an enjoyable > computer game that > collects explanatory descriptions of images. > People play the game > because it is fun, and as a side effect of game > play we collect > valuable information. Given any image from the > World Wide Web, > Phetch can output a correct annotation for it. > The collected data > can be applied towards significantly improving > Web accessibility. In > addition to improving accessibility, Phetch is > an example of a new > class of games that provide entertainment in > exchange for human > processing power. In essence, we solve a typical > computer vision > problem with HCI tools alone. >