Padding-free Convolution based on Preservation of Differential Characteristics of Kernels

TL;DR

This paper introduces a padding-free convolution method that preserves differential characteristics of kernels, avoiding boundary effects and maintaining image size without additional padding or interpolation.

Contribution

The paper proposes a novel, mathematically rigorous convolution technique that eliminates the need for padding by transforming incomplete window convolution into a local differential operator.

Findings

Superior performance in physics field filtering

Improved results in CNN image tasks

No restrictions on image or kernel sizes

Abstract

Convolution is a fundamental operation in image processing and machine learning. Aimed primarily at maintaining image size, padding is a key ingredient of convolution, which, however, can introduce undesirable boundary effects. We present a non-padding-based method for size-keeping convolution based on the preservation of differential characteristics of kernels. The main idea is to make convolution over an incomplete sliding window "collapse" to a linear differential operator evaluated locally at its central pixel, which no longer requires information from the neighbouring missing pixels. While the underlying theory is rigorous, our final formula turns out to be simple: the convolution over an incomplete window is achieved by convolving its nearest complete window with a transformed kernel. This formula is computationally lightweight, involving neither interpolation or extrapolation nor restrictions on image and kernel sizes. Our method favours data with smooth boundaries, such as high-resolution images and fields from physics. Our experiments include: i) filtering analytical and non-analytical fields from computational physics and, ii) training convolutional neural networks (CNNs) for the tasks of image classification, semantic segmentation and super-resolution reconstruction. In all these experiments, our method has exhibited visible superiority over the compared ones.