Sunday December 20, 2015

What your yearbook photo says about social evolution

Discriminative clusters of high school girls’ styles from the 1970's. The left-most entry in each row displays the cluster average.

Discriminative clusters of high school girls’ styles from the 1970's. The left-most entry in each row displays the cluster average. (Shiry Ginosar)

Listen 9:47

A team of computer scientists has used deep learning algorithms on thousands of yearbook photos to perform a scalable historical analysis of large collections of images. Shiry Ginosar explains how these photos have helped us gain new insights into everything from fashion to the evolution of the smile.  

You can find most of the digitized American yearbooks at the Internet Archive.


Images via A Century of Portraits: A Visual Historical Record of American High School Yearbooks

304 yearbook1

Average images of high school students for each decade of the 20th century.

304 yearbook2

The left-most entry in each row displays the cluster average. The clusters correspond to the quintessential hair and accessory styles of each decade. These fashions emerge from the data in a weakly-supervised, data-driven process.

304 yearbook3

Smiles increasing over time, but women always smile more than men.