Finding faces in Google Maps terrain
Something our human eyes seem to do, without any prompting, is to pick out shapes and structures that resemble other shapes and structures. Called pareidolia, it's a form of pattern recognition -- and a good example is the way we often see a human face where only a random collection of shapes or shadows exists. This, it is now known, is the reason for the infamousface on Mars.
Our own Earth, as folded and rippled as it is, is also prone to this phenomenon when viewed from above: the Badlands Guardian, discovered on Google Earth in 2006, for example. But we're sure there are many more human-esque faces lurking in strange corners of the Earth.
That is the premise behind Google Faces, a project by Berlin design studio Onformative: can pareidolia be imitated by a machine? Using OpenFrameworks, the studio has created an application that crawls Google Maps, using facial recognition algorithms to seek out areas that look like faces.
The application uses a virtual browser to search Google Maps, transferring data back to the standalone application using ofxBerkelium to capture and store images of any "faces" found, communicating via Javascript. When the application has crawled all the available images, it jumps to the next zoom level and starts all over again.
Already, the program has been around the world several times, and the ground to cover only gets bigger as it zooms in farther.
"As it continues to travel the world within the upcoming months, it continuously zooms into the earth," Onformative said on the project page. "This process decreases the step-size for each iteration and therefore increases the amount of images and travel time exponentially. Some of the detected images aren't usable at all, as we are not able to recognize any face-like patterns within the detected images. Other satellite images, on the other hand, inspired our imagination in a tremendous, yet funny way."
0 comments:
Post a Comment