Can the early human visual system compete with Deep Neural Networks?

TL;DR

This study compares early human visual processing with deep neural networks on distorted images, revealing humans outperform DNNs under challenging conditions within a limited viewing time.

Contribution

It introduces a novel experimental setup focusing on early human visual mechanisms and demonstrates their superior robustness over current deep neural networks.

Findings

Humans outperform DNNs on blurry images.

Humans outperform DNNs on noisy images.

Early human visual processing is more robust than DNNs.

Abstract

We study and compare the human visual system and state-of-the-art deep neural networks on classification of distorted images. Different from previous works, we limit the display time to 100ms to test only the early mechanisms of the human visual system, without allowing time for any eye movements or other higher level processes. Our findings show that the human visual system still outperforms modern deep neural networks under blurry and noisy images. These findings motivate future research into developing more robust deep networks.