Ulicny, Matej and Krylov, Vladimir A. and Dahyot, Rozenn (2022) Harmonic convolutional networks based on discrete cosine transform. Pattern Recognition, 129 (108707). pp. 1-12. ISSN 00313203

Preview

Download (2MB) | Preview

Abstract

Convolutional neural networks (CNNs) learn filters in order to capture local correlation patterns in feature space. We propose to learn these filters as combinations of preset spectral filters defined by the Discrete Cosine Transform (DCT). Our proposed DCT-based harmonic blocks replace conventional convolutional layers to produce partially or fully harmonic versions of new or existing CNN architectures. Using DCT energy compaction properties, we demonstrate how the harmonic networks can be efficiently compressed by truncating high-frequency information in harmonic blocks thanks to the redundancies in the spectral domain. We report extensive experimental validation demonstrating benefits of the introduction of harmonic blocks into state-of-the-art CNN models in image classification, object detection and semantic segmentation applications.

Item Type:	Article
Keywords:	Harmonic network; Convolutional neural network; Discrete cosine transform; Image classification; Object detection; Semantic segmentation;
Academic Unit:	Faculty of Science and Engineering > Computer Science Faculty of Science and Engineering > Research Institutes > Hamilton Institute
Item ID:	17749
Identification Number:	https://doi.org/10.1016/j.patcog.2022.108707
Depositing User:	Rozenn Dahyot
Date Deposited:	25 Oct 2023 14:33
Journal or Publication Title:	Pattern Recognition
Publisher:	Elsevier
Refereed:	Yes
URI:	Publisher
Use Licence:	This item is available under a Creative Commons Attribution Non Commercial Share Alike Licence (CC BY-NC-SA). Details of this licence are available here

Repository Staff Only(login required)

Item control page

Downloads

Origin of downloads