How is Contrast Encoded in Deep Neural Networks?

TL;DR

This paper investigates how deep neural networks encode contrast, revealing that certain architectural features like multiple convolutional layers before pooling help mitigate contrast variations, with parallels to biological vision.

Contribution

The study uncovers the role of specific convolutional layer arrangements in contrast tolerance and draws biological parallels, advancing understanding of DNN visual processing.

Findings

Networks with more than one convolutional layer before first max-pooling are more contrast tolerant.

The last convolutional layer before pooling acts as a contrast variation mitigator.

Similarities between DNN mechanisms and biological visual systems were observed.

Abstract

Contrast is a crucial factor in visual information processing. It is desired for a visual system - irrespective of being biological or artificial - to "perceive" the world robustly under large potential changes in illumination. In this work, we studied the responses of deep neural networks (DNN) to identical images at different levels of contrast. We analysed the activation of kernels in the convolutional layers of eight prominent networks with distinct architectures (e.g. VGG and Inception). The results of our experiments indicate that those networks with a higher tolerance to alteration of contrast have more than one convolutional layer prior to the first max-pooling operator. It appears that the last convolutional layer before the first max-pooling acts as a mitigator of contrast variation in input images. In our investigation, interestingly, we observed many similarities between the mechanisms of these DNNs and biological visual systems. These comparisons allow us to understand more profoundly the underlying mechanisms of a visual system that is grounded on the basis of "data-analysis".