Deep Convolutional Framelets: A General Deep Learning Framework for Inverse Problems

By Jong Chul Ye, Yoseob Han, E. Cha. SIAM Journal of Imaging Sciences, 2017

328 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computational Mechanics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/5ad0521b0ed5b144e4cb7ac4c14d45aa42ebd98c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Jong Chul Ye
2. Yoseob Han
3. Jaejun Yoo
4. Won Chang
5. Eunhee Kang
6. Sungho Tak
7. Mingxia Liu
8. E. Cha

In Computational Mechanics

1. Jaejun Yoo
2. Jong Chul Ye
3. Yoseob Han
4. Jung-Woo Ha

In Computer Vision and Pattern Recognition

1. Jong Chul Ye
2. Jaejun Yoo
3. Jung-Woo Ha
4. Youngjung Uh
5. Yoseob Han
6. Sanghyuk Chun
7. Luc Van Gool
8. Dacheng Tao

Citations in the corpus

• Deep Learning on Image Denoising: An overview (2019, unknown): “Inspired by a low-rank Hankel matrix in low-level vision, Ye et al. [242] provided convolution frames to explain the connection between signal processing and deep learning by convolving local and nonlocal bases.”
• Framing U-Net via Deep Convolutional Framelets: Application to Sparse-View CT (2017, methodology): “or a given multi-channel signal F:= [f 1 f p] 2Rn p ; (10) an extended Hankel matrix is constructed by stacking Hankel matrices side by side: H dj p(F) := H d(f 1) H (f) H d(f) : (11) As explained in [33], the Hankel matrix is closely related to the convolution operations in CNN. Specifically, for a ”
• Photorealistic Style Transfer via Wavelet Transforms (2019, background): “) It allows WCT(2) to fully reconstruct the signal without any post-processing steps, thanks to the favorable properties of wavelets providing minimal information loss [32, 33].”
• Deep Residual Learning for Accelerated MRI Using Magnitude and Phase Networks (2018, background): “ory of deep convolutional framelets [32], which was the main motivation for (7).”
• Deep Magnetic Resonance Image Reconstruction: Inverse Problems Meet Neural Networks (2020, unknown): “In [19], Ye et al.”
• Deep Learning Techniques for Inverse Problems in Imaging (2020, unknown): “architectures inspired by wavelet decompositions [79].”
• A Gentle Introduction to Deep Learning in Medical Image Processing (2018, unknown): “embed a multi-scale transform into the encoder and decoder of a U-net-like network, which gives rise to the concept of deep convolutional framelets [108].”
• Deep Convolutional Framelet Denosing for Low-Dose CT via Wavelet Residual Network (2017, methodology): “Here, the theory of the deep convolutional framelet [25] is briefly reviewed to make this work self-contained.”