Frequency-Adaptive Dilated Convolution for Semantic Segmentation

TL;DR

This paper introduces Frequency-Adaptive Dilated Convolution (FADC), a novel method that dynamically adjusts dilation rates based on local frequency information, enhancing receptive fields and bandwidth in semantic segmentation tasks.

Contribution

It proposes FADC and two plug-in modules, AdaKern and FreqSelect, to improve dilated convolution by spectrum analysis, enabling adaptive receptive fields and bandwidth control.

Findings

Improved segmentation accuracy on benchmark datasets.

Enhanced receptive field size and effective bandwidth.

Validated effectiveness through extensive experiments.

Abstract

Dilated convolution, which expands the receptive field by inserting gaps between its consecutive elements, is widely employed in computer vision. In this study, we propose three strategies to improve individual phases of dilated convolution from the view of spectrum analysis. Departing from the conventional practice of fixing a global dilation rate as a hyperparameter, we introduce Frequency-Adaptive Dilated Convolution (FADC), which dynamically adjusts dilation rates spatially based on local frequency components. Subsequently, we design two plug-in modules to directly enhance effective bandwidth and receptive field size. The Adaptive Kernel (AdaKern) module decomposes convolution weights into low-frequency and high-frequency components, dynamically adjusting the ratio between these components on a per-channel basis. By increasing the high-frequency part of convolution weights, AdaKern captures more high-frequency components, thereby improving effective bandwidth. The Frequency Selection (FreqSelect) module optimally balances high- and low-frequency components in feature representations through spatially variant reweighting. It suppresses high frequencies in the background to encourage FADC to learn a larger dilation, thereby increasing the receptive field for an expanded scope. Extensive experiments on segmentation and object detection consistently validate the efficacy of our approach. The code is publicly available at https://github.com/Linwei-Chen/FADC.