HFDDLap: High-Low Frequency Differentiation Dynamic Laplacian Pyramid Network for Image Super-Resolution.

Saved in:

Bibliographic Details
Title:	HFDDLap: High-Low Frequency Differentiation Dynamic Laplacian Pyramid Network for Image Super-Resolution.
Authors:	LIU, Peituan¹ liupeituan@163.com, LI, Tie¹ tielitg@163.com, LI, Rui¹ lr_1696@126.com, SONG, Bo¹ 1452254876@qq.com
Source:	Technical Gazette / Tehnički Vjesnik. 2026, Vol. 33 Issue 1, p391-404. 14p.
Subjects:	Convolutional neural networks, Deconvolution (Mathematics), High resolution imaging, Image processing
Abstract:	Existing single image super-resolution (SR) algorithms based on convolution neural networks (CNNs) have achieved usable visual results. However, they often encounter artifacts and blurring, particularly at large scaling factors (e.g., 4×, 8×), due to significant loss of high-frequency information. To address these challenges, we propose a novel high-low frequency differentiation dynamic Laplacian pyramid network (HFDDLap). Our approach introduces a learnable high-low frequency differentiation convolution (HLC) within high-low frequency differentiation residual channel attention blocks (HL-RCAB) to effectively capture and differentiate high- and low-frequency components, enhancing detail preservation. Additionally, we employ a dynamic deconvolution (DDC) that adaptively generates upsampling kernels based on input features, improving reconstruction accuracy by reducing feature distortion. Extensive experiments on 4× and 8× SR demonstrate that our proposed method effectively reconstructs edge details, produces satisfactory SR results and outperforms some state-of-the-art (SOTA) methods in terms of evaluation metrics. [ABSTRACT FROM AUTHOR]
	Copyright of Technical Gazette / Tehnički Vjesnik is the property of Tehnicki Vjesnik and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Existing single image super-resolution (SR) algorithms based on convolution neural networks (CNNs) have achieved usable visual results. However, they often encounter artifacts and blurring, particularly at large scaling factors (e.g., 4×, 8×), due to significant loss of high-frequency information. To address these challenges, we propose a novel high-low frequency differentiation dynamic Laplacian pyramid network (HFDDLap). Our approach introduces a learnable high-low frequency differentiation convolution (HLC) within high-low frequency differentiation residual channel attention blocks (HL-RCAB) to effectively capture and differentiate high- and low-frequency components, enhancing detail preservation. Additionally, we employ a dynamic deconvolution (DDC) that adaptively generates upsampling kernels based on input features, improving reconstruction accuracy by reducing feature distortion. Extensive experiments on 4× and 8× SR demonstrate that our proposed method effectively reconstructs edge details, produces satisfactory SR results and outperforms some state-of-the-art (SOTA) methods in terms of evaluation metrics. [ABSTRACT FROM AUTHOR]
ISSN:	13303651
DOI:	10.17559/TV-20250204002330