Visual Illusions Also Deceive Convolutional Neural Networks: Analysis and Implications

TL;DR

This study demonstrates that convolutional neural networks are susceptible to visual illusions similar to humans, but their responses can differ significantly, highlighting both parallels and differences in visual perception between artificial and biological systems.

Contribution

The paper reveals that CNNs trained on natural images are deceived by brightness and color illusions, showing similarities to human perception but also notable differences.

Findings

CNNs respond to illusions similarly to humans in some aspects

Responses of CNNs can significantly differ from human perception

Implications for using CNNs to model human visual processing

Abstract

Visual illusions allow researchers to devise and test new models of visual perception. Here we show that artificial neural networks trained for basic visual tasks in natural images are deceived by brightness and color illusions, having a response that is qualitatively very similar to the human achromatic and chromatic contrast sensitivity functions, and consistent with natural image statistics. We also show that, while these artificial networks are deceived by illusions, their response might be significantly different to that of humans. Our results suggest that low-level illusions appear in any system that has to perform basic visual tasks in natural environments, in line with error minimization explanations of visual function, and they also imply a word of caution on using artificial networks to study human vision, as previously suggested in other contexts in the vision science literature.