Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations

Hagay Michaeli, Tomer Michaeli, Daniel Soudry

CVPR 2023 pp. 16333-16342

doi:10.1109/CVPR52729.2023.01567 /cvpr/2023/michaeli2023cvpr-aliasfree/

Abstract

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 down-sampling layers. The existing architectural solutions to prevent the aliasing effects are partial since they do not solve those effects that originate in non-linearities. We propose an extended anti-aliasing method that tackles both down-sampling 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.

PDF CVPR Semantic Scholar

Cite

Text

Michaeli et al. "Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations." Conference on Computer Vision and Pattern Recognition, 2023. doi:10.1109/CVPR52729.2023.01567

Markdown

[Michaeli et al. "Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations." Conference on Computer Vision and Pattern Recognition, 2023.](https://mlanthology.org/cvpr/2023/michaeli2023cvpr-aliasfree/) doi:10.1109/CVPR52729.2023.01567

BibTeX

@inproceedings{michaeli2023cvpr-aliasfree,
  title     = {{Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations}},
  author    = {Michaeli, Hagay and Michaeli, Tomer and Soudry, Daniel},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2023},
  pages     = {16333-16342},
  doi       = {10.1109/CVPR52729.2023.01567},
  url       = {https://mlanthology.org/cvpr/2023/michaeli2023cvpr-aliasfree/}
}