Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Michaeli, Hagay, Michaeli, Tomer, Soudry, Daniel
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914109735305216
author Michaeli, Hagay
Michaeli, Tomer
Soudry, Daniel
author_facet Michaeli, Hagay
Michaeli, Tomer
Soudry, Daniel
contents Although CNNs are believed to be invariant to translations, recent works have shown this is not the case, due to aliasing effects that stem from downsampling layers. The existing architectural solutions to prevent aliasing are partial since they do not solve these effects, that originate in non-linearities. We propose an extended anti-aliasing method that tackles both downsampling and non-linear layers, thus creating truly alias-free, shift-invariant CNNs. We show that the presented model is invariant to integer as well as fractional (i.e., sub-pixel) translations, thus outperforming other shift-invariant methods in terms of robustness to adversarial translations.
format Preprint
id arxiv_https___arxiv_org_abs_2303_08085
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations
Michaeli, Hagay
Michaeli, Tomer
Soudry, Daniel
Computer Vision and Pattern Recognition
Image and Video Processing
Although CNNs are believed to be invariant to translations, recent works have shown this is not the case, due to aliasing effects that stem from downsampling layers. The existing architectural solutions to prevent aliasing are partial since they do not solve these effects, that originate in non-linearities. We propose an extended anti-aliasing method that tackles both downsampling and non-linear layers, thus creating truly alias-free, shift-invariant CNNs. We show that the presented model is invariant to integer as well as fractional (i.e., sub-pixel) translations, thus outperforming other shift-invariant methods in terms of robustness to adversarial translations.
title Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2303.08085