• Login
    View Item 
    •   NWU-IR Home
    • Research Output
    • Faculty of Engineering
    • View Item
    •   NWU-IR Home
    • Research Output
    • Faculty of Engineering
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Tracking translation invariance in CNNs

    Thumbnail
    View/Open
    TrackingTranslationInvariance_Myburgh.pdf (346.9Kb)
    Date
    2020
    Author
    Myburgh, Johannes C.
    Mouton, Coenraad
    Davel, Marelie H.
    Metadata
    Show full item record
    Abstract
    Although Convolutional Neural Networks (CNNs) are widely used, their translation invariance (ability to deal with translated inputs) is still subject to some controversy. We explore this question using translation-sensitivity maps to quantify how sensitive a standard CNN is to a translated input. We propose the use of cosine similarity as sensitivity metric over Euclidean distance, and discuss the importance of restricting the dimensionality of either of these metrics when comparing architectures. Our main focus is to investigate the effect of different architectural components of a standard CNN on that network’s sensitivity to translation. By varying convolutional kernel sizes and amounts of zero padding, we control the size of the feature maps produced, allowing us to quantify the extent to which these elements influence translation invariance. We also measure translation invariance at different locations within the CNN to determine the extent to which convolutional and fully connected layers, respectively, contribute to the translation invariance of a CNN as a whole. Our analysis indicates that both convolutional kernel size and feature map size have a systematic influence on translation invariance. We also see that convolutional layers contribute less than expected to translation invariance, when not specifically forced to do so.
    URI
    http://hdl.handle.net/10394/36935
    https://doi.org/10.1007/978-3-030-66151-9_18
    https://arxiv.org/abs/2104.05997
    Collections
    • Faculty of Engineering [1136]

    Copyright © North-West University
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Browse

    All of NWU-IR Communities & CollectionsBy Issue DateAuthorsTitlesSubjectsAdvisor/SupervisorThesis TypeThis CollectionBy Issue DateAuthorsTitlesSubjectsAdvisor/SupervisorThesis Type

    My Account

    LoginRegister

    Copyright © North-West University
    Contact Us | Send Feedback
    Theme by 
    Atmire NV