Capacity limitations of visual search in deep convolutional neural networks

2017/07/31 by Endel Põder, Endel Poder, Poder, Endel
Biochemistry, Genetics and Molecular Biology · Computer Science · Neuroscience · #Face Recognition and Perception #Neural dynamics and brain function #Visual perception and processing mechanisms #cs.CV #q-bio.NC

paper · pdf · doi:10.48550/arxiv.1707.09775

10 pages, 4 figures

arxiv created 2021/03/29 · arxiv updated 2021/03/30

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.

Related

• Comparing object recognition in humans and deep convolutional neural networks -- An eye tracking study
• Comparing deep neural networks against humans: object recognition when the signal gets weaker
• CNN-based search model underestimates attention guidance by simple visual features
• Seeing eye-to-eye? A comparison of object recognition performance in humans and deep convolutional neural networks under image manipulation
• Crowding in humans is unlike that in convolutional neural networks