Padding-free Convolution based on Preservation of Differential Characteristics of Kernels
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.
Code (0)
등록된 구현이 없습니다.
Tasks
image-classificationImage ClassificationSemantic SegmentationSuper-ResolutionMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Deep Convolutional Neural Networks with Zero-Padding: Feature Extraction and Learning
This paper studies the performance of deep convolutional neural networks (DCNNs) with zero-padding in feature extraction and learning. After verifying the roles of zero-padding in enabling translation-equivalence, and po…
TranslationUnveiling The Mask of Position-Information Pattern Through the Mist of Image Features
Recent studies show that paddings in convolutional neural networks encode absolute position information which can negatively affect the model performance for certain tasks. However, existing metrics for quantifying the s…
PositionRandom Padding Data Augmentation
The convolutional neural network (CNN) learns the same object in different positions in images, which can improve the recognition accuracy of the model. An implication of this is that CNN may know where the object is. Th…
Data Augmentationimage-classificationImage ClassificationPositionPartial Convolution based Padding
In this paper, we present a simple yet effective padding scheme that can be used as a drop-in module for existing convolutional neural networks. We call it partial convolution based padding, with the intuition that the p…
General ClassificationSemantic SegmentationOn the Interplay of Convolutional Padding and Adversarial Robustness
It is common practice to apply padding prior to convolution operations to preserve the resolution of feature-maps in Convolutional Neural Networks (CNN). While many alternatives exist, this is often achieved by adding a …
Adversarial Robustness