期刊
IEEE SIGNAL PROCESSING LETTERS
卷 29, 期 -, 页码 1943-1947出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LSP.2022.3205275
关键词
Sensors; Generators; Decoding; Training; Image reconstruction; Image coding; Loss measurement; Information bottleneck; image compressed sens- ing; deep learning
资金
- Korea Institute of Civil Engineering and Building Technology (KICT) [20220238-001]
- National Research Council of Science & Technology (NST), Republic of Korea [20220238-001] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)
This paper proposes a novel training process that simultaneously learns the sensing and decoder networks using Information Bottleneck theory, to improve the reconstruction performance of image compressed sensing algorithms.
Image Compressed Sensing (CS) has achieved a lot of performance improvement thanks to advances in deep networks. The CS method is generally composed of a sensing and a decoder. The sensing and decoder networks have a significant impact on the reconstruction performance, and it is obvious that both two networks must be in harmony. However, previous studies have focused on designing the loss function considering only the decoder network. In this paper, we propose a novel training process that can learn sensing and decoder networks simultaneously using Information Bottleneck (IB) theory. By maximizing importance through proposed importance generator, the sensing network is trained to compress important information for image reconstruction of the decoder network. The representative experimental results demonstrate that the proposed method is applied in recently proposed CS algorithms and increases the reconstruction performance with large margin in all CS ratios.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据