TY - JOUR

T1 - One-Two-One Network for Compression Artifacts Reduction in Remote Sensing

AU - Zhang, Baochang

AU - Gu, Jiaxin

AU - Chen, Chen

AU - Han, Jungong

AU - Su, Xiangbo

AU - Cao, Xianbin

AU - Liu, Jianzhuang

N1 - s is the author’s version of a work that was accepted for publication in ISPRS Journal of Photogrammetry and Remote Sensing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in ISPRS Journal of Photogrammetry and Remote Sensing, 145, Part A, 2018 DOI: 10.1016/j.isprsjprs.2018.01.003

PY - 2018/11

Y1 - 2018/11

N2 - Compression artifacts reduction (CAR) is a challenging problem in the field of remote sensing. Most recent deep learning based methods have demonstrated superior performance over the previous hand-crafted methods. In this paper, we propose an end-to-end one-two-one (OTO) network, to combine different deep models, i.e., summation and difference models, to solve the CAR problem. Particularly, the difference model motivated by the Laplacian pyramid is designed to obtain the high frequency information, while the summation model aggregates the low frequency information. We provide an in-depth investigation into our OTO architecture based on the Taylor expansion, which shows that these two kinds of information can be fused in a nonlinear scheme to gain more capacity of handling complicated image compression artifacts, especially the blocking effect in compression. Extensive experiments are conducted to demonstrate the superior performance of the OTO networks, as compared to the state-of-the-arts on remote sensing datasets and other benchmark datasets. The source code will be available here: https://github.com/bczhangbczhang/.

AB - Compression artifacts reduction (CAR) is a challenging problem in the field of remote sensing. Most recent deep learning based methods have demonstrated superior performance over the previous hand-crafted methods. In this paper, we propose an end-to-end one-two-one (OTO) network, to combine different deep models, i.e., summation and difference models, to solve the CAR problem. Particularly, the difference model motivated by the Laplacian pyramid is designed to obtain the high frequency information, while the summation model aggregates the low frequency information. We provide an in-depth investigation into our OTO architecture based on the Taylor expansion, which shows that these two kinds of information can be fused in a nonlinear scheme to gain more capacity of handling complicated image compression artifacts, especially the blocking effect in compression. Extensive experiments are conducted to demonstrate the superior performance of the OTO networks, as compared to the state-of-the-arts on remote sensing datasets and other benchmark datasets. The source code will be available here: https://github.com/bczhangbczhang/.

KW - Compression artifacts reduction

KW - Remote sensing

KW - Deep learning

KW - One-two-one network

U2 - 10.1016/j.isprsjprs.2018.01.003

DO - 10.1016/j.isprsjprs.2018.01.003

M3 - Journal article

VL - 145

SP - 184

EP - 196

JO - ISPRS Journal of Photogrammetry and Remote Sensing

JF - ISPRS Journal of Photogrammetry and Remote Sensing

SN - 0924-2716

IS - Part A

ER -