Noise increases the correspondence between artificial and human vision.

Journal: PLoS biology
Published Date:

Abstract

The best performing computer vision systems are based on deep neural networks (DNNs). A study in this issue of PLOS Biology shows that DNNs trained on noisy stimuli are better than standard DNNs at mirroring both human behavioral and neural visual responses.

Authors

  • Jessica A F Thompson
    Human Information Processing Lab, Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom.