This visualization uses PixPlot, software developed at Yale’s Digital Humanities Lab that sorts images using an unsupervised clustering algorithm, allowing viewers to explore visual relationships between images that may not immediately be apparent in large collections.
Here we use PixPlot to visualize 8,789 faces sampled from 100 issues of our Time magazine archive, spanning the years 1923 to 2014. You can explore clusters, pan, zoom, sort, or click on individual images to reveal more information about the context of each image.
Select a hotspot to explore visually similar clusters.
Lasso to select. Shift + drag to edit selection.
Use two fingers to zoom in and out.