Capacity limitations of visual search in deep convolutional neural networks

TL;DR

This paper investigates the capacity limitations of deep convolutional neural networks in visual search tasks, revealing that they exhibit similar capacity constraints as human vision, despite architectural differences.

Contribution

It provides a comparative analysis showing that neural networks share human-like capacity limitations in visual search, which was previously underexplored.

Findings

Neural networks show no clear difference between simple feature searches and complex configurations.

Both simple and complex visual searches reveal similar capacity limitations in neural networks.

Neural networks' performance differs qualitatively from human visual search behavior.

Abstract

Deep convolutional neural networks follow roughly the architecture of biological visual systems and have shown a performance comparable to human observers in object recognition tasks. In this study, I tested three pretrained deep neural networks in visual search for simple visual features, and for feature configurations. The results reveal a qualitative difference from human performance. It appears that there is no clear difference between searches for simple features that pop out in experiments with humans, and for feature configurations that exhibit strict capacity limitations in human vision. Both types of stimuli reveal comparable capacity limitations in the neural networks tested here.