Diptychs of human and machine perceptions

Abstract

We propose visual creations that put differences in algorithms and humans \emph{perceptions} into perspective. We exploit saliency maps of neural networks and visual focus of humans to create diptychs that are reinterpretations of an original image according to both machine and human attentions. Using those diptychs as a qualitative evaluation of perception, we discuss some crucial issues of current \textit{task-oriented} artificial intelligence.

arXiv comment7 pages, 36 images

References in corpus (6)

• CAT2000: A Large Scale Fixation Dataset for Boosting Saliency Research
• Creative GANs for generating poems, lyrics, and metaphors
• Dialog on a canvas with a machine
• Mechanisms of Artistic Creativity in Deep Learning Neural Networks
• Interactive Neural Style Transfer with Artists
• Predicting A Creator's Preferences In, and From, Interactive Generative Art