Volume 1, Issue 3
Model Meets Deep Learning in Image Inverse Problems

Na Wang & Jian Sun

CSIAM Trans. Appl. Math., 1 (2020), pp. 365-386.

Published online: 2020-09

Preview Full PDF 214 2009
Export citation
  • Abstract

Image inverse problem aims to reconstruct or restore high-quality images from observed samples or degraded images, with wide applications in imaging sciences. The traditional methods rely on mathematical models to invert the process of image sensing or degradation. But these methods require good design of image prior or regularizer that is hard to be hand-crafted. In recent years, deep learning has been introduced to image inverse problems by learning to invert image sensing or degradation process. In this paper, we will review a new trend of methods for image inverse problem that combines the imaging/degradation model with deep learning approach. These methods are typically designed by unrolling some optimization algorithms or statistical inference algorithms into deep neural networks. The ideas combining deep learning and models are also emerging in other fields such as PDE, control, etc. We will also summarize and present perspectives along this research direction.

  • Keywords

Image inverse problem, model-driven deep learning, statistical model, optimization model.

  • AMS Subject Headings

00-XX

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{CSIAM-AM-1-365, author = {Na Wang , and Jian Sun , }, title = {Model Meets Deep Learning in Image Inverse Problems}, journal = {CSIAM Transactions on Applied Mathematics}, year = {2020}, volume = {1}, number = {3}, pages = {365--386}, abstract = {

Image inverse problem aims to reconstruct or restore high-quality images from observed samples or degraded images, with wide applications in imaging sciences. The traditional methods rely on mathematical models to invert the process of image sensing or degradation. But these methods require good design of image prior or regularizer that is hard to be hand-crafted. In recent years, deep learning has been introduced to image inverse problems by learning to invert image sensing or degradation process. In this paper, we will review a new trend of methods for image inverse problem that combines the imaging/degradation model with deep learning approach. These methods are typically designed by unrolling some optimization algorithms or statistical inference algorithms into deep neural networks. The ideas combining deep learning and models are also emerging in other fields such as PDE, control, etc. We will also summarize and present perspectives along this research direction.

}, issn = {2708-0579}, doi = {https://doi.org/10.4208/csiam-am.2020-0016}, url = {http://global-sci.org/intro/article_detail/csiam-am/18301.html} }
TY - JOUR T1 - Model Meets Deep Learning in Image Inverse Problems AU - Na Wang , AU - Jian Sun , JO - CSIAM Transactions on Applied Mathematics VL - 3 SP - 365 EP - 386 PY - 2020 DA - 2020/09 SN - 1 DO - http://doi.org/10.4208/csiam-am.2020-0016 UR - https://global-sci.org/intro/article_detail/csiam-am/18301.html KW - Image inverse problem, model-driven deep learning, statistical model, optimization model. AB -

Image inverse problem aims to reconstruct or restore high-quality images from observed samples or degraded images, with wide applications in imaging sciences. The traditional methods rely on mathematical models to invert the process of image sensing or degradation. But these methods require good design of image prior or regularizer that is hard to be hand-crafted. In recent years, deep learning has been introduced to image inverse problems by learning to invert image sensing or degradation process. In this paper, we will review a new trend of methods for image inverse problem that combines the imaging/degradation model with deep learning approach. These methods are typically designed by unrolling some optimization algorithms or statistical inference algorithms into deep neural networks. The ideas combining deep learning and models are also emerging in other fields such as PDE, control, etc. We will also summarize and present perspectives along this research direction.

Na Wang & Jian Sun. (2020). Model Meets Deep Learning in Image Inverse Problems. CSIAM Transactions on Applied Mathematics. 1 (3). 365-386. doi:10.4208/csiam-am.2020-0016
Copy to clipboard
The citation has been copied to your clipboard