Revisiting Edge Detection in Convolutional Neural Networks

Minh Le, Subhradeep Kayal

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    Abstract

    The ability to detect edges is a fundamental attribute necessary to truly capture visual concepts. In this paper, we prove that edges cannot be represented properly in the first convolutional layer of a neural network, and further show that they are poorly captured in popular neural network architectures such as VGG-16 and ResNet. The neural networks are found to rely on color information, which might vary in unexpected ways outside of the datasets used for their evaluation. To improve their robustness, we propose edge-detection units and show that they reduce performance loss and generate qualitatively different representations. By comparing various models, we show that the robustness of edge detection is an important factor contributing to the robustness of models against color noise.
    Original languageAmerican English
    Title of host publicationInternational Joint Conference on Neural Networks
    StatePublished - 2021

    Fingerprint Dive into the research topics of 'Revisiting Edge Detection in Convolutional Neural Networks'. Together they form a unique fingerprint.

    Cite this