Luca Damasco

05 Apr 2016

Teenie Harris was a professional photographer in Pittsburgh’s Hill District from the 1930’s through the 1970’s. His work catalogued the lives of Pittsburgh’s African American community for generations. After his death in the late 1990’s, Teenie left behind an archive of over 60,000 photographs leaving the insurmountable task of cataloguing them all to the art historians of the Carnegie Museum of Art.

Initially, I was intrigued by the shear number of people depicted in Teenie’s archive and also by the fact that most of them remained nameless. Using classic computer vision techniques, I programmatically removed hundreds of faces from the images to create a quick snapshot of a part of Teenie’s world.

Exploring the images, it was clear that there were pockets of people Teenie loved to photograph at different times in his life, but there was no simple way to find images which connected individuals in each of the photos. To solve this problem, the team uploaded a small portion of the archive to Google Photos, which began to programmatically sort images by categories as well as the people in them. Here are a few of the Tags and connections it found.

Using this method of sorting, historians will be able to do hours of work in seconds with a single search. The archive will soon be sorted to connect each of the photo’s individual reference numbers with their associated tags.