A picture can paint a thousand words. But isn’t it better if your camera can actually describe images for you? That is now possible with the Descriptive Camera. The camera is actually a prototype developed by Matt Richardson, a computer programmer and photographer rolled into one. He probably made the prototype for a Computational Camera class at the New York University where he is currently taking his Master’s Degree.
According to the inventor, the Descriptive Camera works by snapping pictures first then sending those images to the Mechanical Turk of Amazon where a couple of people there are paid to do jobs that are too complex for computers to solve. That makes sense, after all, today’s computers cannot describe the contents of a picture – proof that we humans are supreme. Anyway, once it’s submitted to the Mechanical Turk, it will be processed there (by humans of course) after which the results will be forwarded back to the camera within 6 minutes.
A thermal printer then prints the description via text in the style of a Polaroid print. As crazy as it seems, we think that the camera will be extremely useful, especially to those who are visually impaired or blind. “I wanted to make something that could create descriptions instead of photos, but I think there is a lot of fun that can be had when you compare the output of the camera and what the description says,” Richardson said.