From CNNs to Shift-Invariant Twin Models Based on Complex Wavelets. (arXiv:2212.00394v2 [cs.CV] UPDATED)

We propose a novel antialiasing method to increase shift invariance and
prediction accuracy in convolutional neural networks. Specifically, we replace
the first-layer combination "real-valued convolutions + max pooling"
($\mathbb{R}$Max) by "complex-valued convolutions + modulus" ($\mathbb{C}$Mod),
which is stable to translations. To justify our approach, we claim that
$\mathbb{C}$Mod and $\mathbb{R}$Max produce comparable outputs when the
convolution kernel is band-pass and oriented (Gabor-like filter). In this
context, $\mathbb{C}$Mod can be considered as a stable alternative to
$\mathbb{R}$Max. Thus, prior to antialiasing, …

accuracy arxiv claim cnns combination context convolution convolutional neural networks kernel max modulus networks neural networks novel pooling prediction prior shift