The Descriptive Camera only outputs the metadata about the content
The back-to-basics camera consists of a shutter button, a USB webcam, a small printer and a simple ethernet connection. In short, the user simply takes a shot as normal but instead of an image a text description is produced.
At the heart of the camera is Amazon’s Mechanical Turk API, a service that offers Human Intelligence Tasks (HITs), essentially time consuming mundane online tasks that anyone can earn instant cash from across the world – in-turn saving the developer money. The developer simply sets the guidelines, the price and then waits for a response from someone online looking to earn a quick buck.
In this case the camera sends the image to the Mechanical Turk API and from there an online worker will describe what they see. A yellow LED indicates that the results are still “developing” in a nod to film-based photo technology. With a HIT price of $1.25, results are returned typically within 6 minutes and sometimes as fast as 3 minutes. The thermal printer outputs the resulting text in the style of a polaroid print. The aim is to challenge us to analyse more closely the context of a photo and attach more information to each image we take.
“The Descriptive Camera works a lot like a regular camera—point it at subject and press the shutter button to capture the scene,” the creator Matt Richardson said. “However, instead of producing an image, this prototype outputs a text description of the scene. Modern digital cameras capture gobs of parsable metadata about photos such as the camera’s settings, the location of the photo, the date, and time, but they don’t output any information about the content of the photo. The Descriptive Camera only outputs the metadata about the content.”
As we amass an incredible amount of photos, it becomes increasingly difficult to manage our collections
The project was launched by Richardson as part of the New York University’s Telecommunications course with the idea that the technology could be used to better manage all our photography collections. It stems from all of us taking more and more photos without necessarily tagging and labeling each snap. The fear is we will all end up hoarding thousands of images that aren’t categorised, that could have easily been automatically referenced.
“As we amass an incredible amount of photos, it becomes increasingly difficult to manage our collections,” Richardson said. “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 is a photo taken by Richardson of an inner-city building – the print-out read: “this is a faded picture of a dilapidated building. It seems to be run down and in need of repairs.”
With cameras already storing metadata, the Descriptive camera was a natural step: “I’ve been thinking a lot about how cameras store a lot of metadata about a photo,” Richardson told BetaBeat. “It’s limited to longitude, latitude, the time of day, the settings, and make and model of the camera. I thought, ‘Hey, wouldn’t it be neat if it also recorded searchable text about what is going on in the picture?’ We have some methods like that: tagging faces algorithmically and finding out who’s who in the picture, but I wanted something more than that.”
Richardson’s invention is still very much its infancy but the idea pushes all of us to think about the long-term picture surrounding each individual image and how we organise our photography collection. Richardson’s vision is simple: “I was picturing a time in which cameras could possibly capture more useful information that can then be searched, cross-referenced and sorted.”