The Descriptive Camera, created by Matt Richardson, “works a lot like a regular camera — point it at subject and press the shutter button to capture the scene. However, instead of producing an image, this prototype outputs a text description of the scene.” Instead of printing the image, the camera sends the information it captures to Amazon’s Mechanical Turk, where humans pick up the job, describe the image and send it back to the camera, a process that takes about three to six minutes. Obviously, right now it’s not the most practical thing in the world, but imagine being able to automatically “tag” photographs with metadata every time you take a picture. It’s like Google’s image search function, only in the real world. Wild. Click through to see photos of the camera, and head here for more information on the project.