Bayesian Deep Learning for Image Reconstruction: From structured sparsity to uncertainty estimation

被引：3

作者：

Dong, Weisheng ^{[1
,2
]}

Wu, Jinjian ^{[3
,4
]}

Li, Leida ^{[5
]}

Shi, Guangming ^{[6
,7
]}

Li, Xin ^{[8
,9
]}

机构：

[1] Hong Kong Polytech Univ, Dept Comp, Hong Kong, Peoples R China

[2] Xidian Univ, Sch Artificial Intelligence, Xian 710071, Peoples R China

[3] Nanyang Technol Univ, Singapore, Singapore

[4] Xidian Univ, Sch Artificial Intelligence, Xian 710071, Peoples R China

[5] Nanyang Technol Univ, Sch Elect & Elect Engn, Rapid Rich Object Search Lab, Singapore, Singapore

[6] Xidian Univ, Sch Elect Engn, Xian, Peoples R China

[7] Xidian Univ, Sch Artificial Intelligence, Xian 710071, Peoples R China

[8] Univ Sci & Technol China, Elect Engn & informat Sci, Hefei, Peoples R China

[9] West Virginia Univ, Lane Dept Comp Sci & Elect Engn, Morgantown, WV 26506 USA

来源：

IEEE SIGNAL PROCESSING MAGAZINE | 2023年 / 40卷 / 01期

基金：

美国国家科学基金会; 国家重点研发计划;

关键词：

Deep learning; Uncertainty; Image coding; Computational modeling; Imaging; Signal processing algorithms; Network architecture; REPRESENTATION; RESTORATION; NETWORK;

D O I：

10.1109/MSP.2022.3176421

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Conventional wisdom in model-based computational imaging incorporates physics-based imaging models, noise characteristics, and image priors into a unified Bayesian framework. Rapid advances in deep learning have inspired a new generation of data-driven computational imaging systems with performances even better than those of their model-based counterparts. However, the design of learning-based algorithms for computational imaging often lacks transparency, making it difficult to optimize the entire imaging system in a complete manner.

引用

页码：73 / 84

页数：12

共 30 条

[1] Solving inverse problems using data-driven models
Arridge, Simon
Maass, Peter
Oktem, Ozan
Schonlieb, Carola-Bibiane
[J]. ACTA NUMERICA, 2019, 28 : 1 - 174
[2] Cheng XJ, 2020, AAAI CONF ARTIF INTE, V34, P10615
[3] Image Super-Resolution Using Deep Convolutional Networks
Dong, Chao
Loy, Chen Change
He, Kaiming
Tang, Xiaoou
[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2016, 38 (02) : 295 - 307
[4] Denoising Prior Driven Deep Neural Network for Image Restoration
Dong, Weisheng
Wang, Peiyao
Yin, Wotao
Shi, Guangming
Wu, Fangfang
Lu, Xiaotong
[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2019, 41 (10) : 2305 - 2318
[5] Nonlocally Centralized Sparse Representation for Image Restoration
Dong, Weisheng
Zhang, Lei
Shi, Guangming
Li, Xin
[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING, 2013, 22 (04) : 1618 - 1628
[6] Sparse Representation Based Image Interpolation With Nonlocal Autoregressive Modeling
Dong, Weisheng
Zhang, Lei
Lukac, Rastislav
Shi, Guangming
[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING, 2013, 22 (04) : 1382 - 1394
[7] Nonlocal Image Restoration With Bilateral Variance Estimation: A Low-Rank Approach
Dong, Weisheng
Shi, Guangming
Li, Xin
[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING, 2013, 22 (02) : 700 - 711
[8] Dong WS, 2011, PROC CVPR IEEE, P457, DOI 10.1109/CVPR.2011.5995478
[9] Deep Gaussian Scale Mixture Prior for Spectral Compressive Imaging
Huang, Tao
Dong, Weisheng
Yuan, Xin
Wu, Jinjian
Shi, Guangming
[J]. 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 16211 - 16220
[10] Jinsun Park, 2020, Computer Vision - ECCV 2020. 16th European Conference. Proceedings. Lecture Notes in Computer Science (LNCS 12358), P120, DOI 10.1007/978-3-030-58601-0_8

← 1 2 3 →