FABNet: Frequency-Aware Binarized Network for Single Image Super-Resolution

作     者：Jiang, Xinrui Wang, Nannan Xin, Jingwei Li, Keyu Yang, Xi Li, Jie Wang, Xiaoyu Gao, Xinbo 

作者机构：Xidian Univ Sch Telecommun Engn State Key Lab Integrated Serv Networks Xian 710071 Shaanxi Peoples R China Chinese Univ Hong Kong Sch Sci & Engn Shenzhen 518172 Peoples R China Chongqing Univ Posts & Telecommun Chongqing Key Lab Image Cognit Chongqing 400065 Peoples R China 

出 版 物：《IEEE TRANSACTIONS ON IMAGE PROCESSING》 (IEEE Trans Image Process)

年 卷 期：2023年第32卷

页      面：6234-6247页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National Key Research and Development Program of China 

主　　题：Quantization (signal) Superresolution Task analysis Neural networks Discrete wavelet transforms Image coding Electronic mail Single image super-resolution binary neural network wavelet decomposition lightweight 

摘      要：Remarkable achievements have been obtained with binary neural networks (BNN) in real-time and energy-efficient single-image super-resolution (SISR) methods. However, existing approaches often adopt the Sign function to quantize image features while ignoring the influence of image spatial frequency. We argue that we can minimize the quantization error by considering different spatial frequency components. To achieve this, we propose a frequency-aware binarized network (FABNet) for single image super-resolution. First, we leverage the wavelet transformation to decompose the features into low-frequency and high-frequency components and then employ a divide-and-conquer strategy to separately process them with well-designed binary network structures. Additionally, we introduce a dynamic binarization process that incorporates learned-threshold binarization during forward propagation and dynamic approximation during backward propagation, effectively addressing the diverse spatial frequency information. Compared to existing methods, our approach is effective in reducing quantization error and recovering image textures. Extensive experiments conducted on four benchmark datasets demonstrate that the proposed methods could surpass state-of-the-art approaches in terms of PSNR and visual quality with significantly reduced computational costs. Our codes are available at https://***/xrjiang527/FABNet-PyTorch.