A picture is worth a thousand words. For the Descriptive Camera by Matt Richardson, this oft-used moniker is actually true–unlike a traditional camera, his camera produces a narrative, not a photo.
The Descriptive Camera, which consists of a USB webcam, a shutter button, a small thermal printer, and an Ethernet connection, captures a photo just like a traditional camera- just point and click to get the desired image. The key to the camera is the Ethernet connection, which sends the captured image to Amazon’s Mechanical Turk API, where workers completing human intelligence tasks (HITs) take a look at the image and then describe what they see.
The process takes between 3-6 minutes and costs roughly $1.25 per image. When a description has been assigned to an image, the thermal printer on the camera then prints out the text, framed by a polaroid-esque photo border.
Richardson designed the Descriptive Camera for his Computational Cameras class at New York University’s Interactive Telecommunications Program (NYU ITP), and it’s a study in how to better manage photo collections:
As we amass an incredible amount of photos, it becomes increasingly difficult to manage our collections. Imagine if descriptive metadata about each photo could be appended to the image on the fly—information about who is in each photo, what they’re doing, and their environment could become incredibly useful in being able to search, filter, and cross-reference our photo collections. Of course, we don’t yet have the technology that makes this a practical proposition, but the Descriptive Camera explores these possibilities.
An example of the Descriptive Camera’s output here, juxtaposed with the image it describes: