Brief description
The artist trained a neural network with faces wondering "what the dreams of a facial recognition system would look like".
The artwork consist of two videos that are a result of this "dreaming" state. The diptych displays the system trained on two different sets of faces.
In one set the images are aligned and cleaned (hence the faces are more clear) in the other set the collected images were quite variant and unaligned (the result more blurry).
Source: https://www.deweyhagborg.com/projects/spurious-memories