Efficient and Interpretable Deep Blind Image Deblurring Via Algorithm Unrolling

Li, Yuelong; Tofighi, Mohammad; Geng, Junyi; Monga, Vishal; Eldar, Yonina C.

doi:10.1109/tci.2020.2964202

Cited by 163 publications

(105 citation statements)

References 69 publications

Supporting

Mentioning

105

Contrasting

Order By: Relevance

“…Blind image restoration [44] [3] [32] aims to directly learn the restoration mapping based on observed samples. However, most existing methods for general natural images are still sensitive to the degradation profile [9] and exhibit poor generalization over unconstrained testing conditions.…”

Section: Blind Face Restorationmentioning

confidence: 99%

HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment

Yang

Liu

Wang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

113

View full text Add to dashboard Cite

Existing face restoration researches typically rely on either the image degradation prior or explicit guidance labels for training, which often lead to limited generalization ability over real-world images with heterogeneous degradation and rich background contents. In this paper, we investigate a more challenging and practical "dual-blind" version of the problem by lifting the requirements on both types of prior, termed as "Face Renovation"(FR). Specifically, we formulate FR as a semantic-guided generation problem and tackle it with a collaborative suppression and replenishment (CSR) approach. This leads to HiFaceGAN, a multi-stage framework containing several nested CSR units that progressively replenish facial details based on the hierarchical semantic guidance extracted from the front-end content-adaptive suppression modules. Extensive experiments on both synthetic and real face images have verified the superior performance of our HiFaceGAN over a wide range of challenging restoration subtasks, demonstrating its versatility, robustness and generalization ability towards real-world face processing applications.

show abstract

Section: Blind Face Restorationmentioning

confidence: 99%

HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment

Yang

Liu

Wang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

113

View full text Add to dashboard Cite

show abstract

“…where A * a , B * a , C * a , D * a , and E * a are individual trainable graph convolutions with filter coefficients that are trainable parameters with subscript a indicating trainable, and NN u (⋅) and NN r (⋅) are two neural networks, which involve trainable parameters. Intuitively, (3a) and (3b) are neural-network implementations of (2a) and (2b), respectively, replacing fixed graph convolutions h * v by trainable graph convolutions A * a ; and (3c) and (3d) are neural-network implementations of the proximal functions (2c) and (2d), respectively, using neural networks to solve sub-optimization problems; see similar substitutions in [14,15,16]. Instead of following the exact mathematical relationship in (2), we allow trainable operators to adaptively learn from data, usually reducing a lot of computation.…”

Section: Algorithm Unrollingmentioning

confidence: 99%

Graph Signal Denoising Via Unrolling Networks

Chen

Eldar

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

We propose an interpretable graph neural network framework to denoise single or multiple noisy graph signals. The proposed graph unrolling networks expand algorithm unrolling to the graph domain and provide an interpretation of the architecture design from a signal processing perspective. We unroll an iterative denoising algorithm by mapping each iteration into a single network layer where the feed-forward process is equivalent to iteratively denoising graph signals. We train the graph unrolling networks through unsupervised learning, where the input noisy graph signals are used to supervise the networks. By leveraging the learning ability of neural networks, we adaptively capture appropriate priors from input noisy graph signals, instead of manually choosing signal priors. To validate the proposed methods, we conduct extensive experiments on both real-world datasets and simulated datasets, and demonstrate that our methods have smaller denoising errors than conventional denoising algorithms and state-of-the-art graph neural networks. For denoising a single smooth graph signal, the normalized mean square error of the proposed networks is around 40% and 60% lower than that of graph Laplacian denoising and graph wavelets, respectively.

show abstract

“…However, deep neural networks are usually designed empirically and their structures lack interpretability, which is a prominent shortcoming. In the past few years, algorithm unrolling/unfolding [12][13][14][15] In this work, we develop a network using algorithm unrolling to localize neurons in tissues from LFM images. The network performs convolutional sparse coding on input epipolar plane images (EPI) [16][17][18], a type of spatio-angular feature constructed from light-field images, and output sparse codes which indicates depth positions.…”

Section: Introductionmentioning

confidence: 99%

Model-Inspired Deep Learning for Light-Field Microscopy with Application to Neuron Localization

Song¹,

Jadan²,

Howe

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Light-field microscopes are able to capture spatial and angular information of incident light rays. This allows reconstructing 3D locations of neurons from a single snap-shot. In this work, we propose a model-inspired deep learning approach to perform fast and robust 3D localization of sources using light-field microscopy images. This is achieved by developing a deep network that efficiently solves a convolutional sparse coding (CSC) problem to map Epipolar Plane Images (EPI) to corresponding sparse codes. The network architecture is designed systematically by unrolling the convolutional Iterative Shrinkage and Thresholding Algorithm (ISTA) while the network parameters are learned from a training dataset. Such principled design enables the deep network to leverage both domain knowledge implied in the model, as well as new parameters learned from the data, thereby combining advantages of model-based and learning-based methods. Practical experiments on localization of mammalian neurons from light-fields show that the proposed approach simultaneously provides enhanced performance, interpretability and efficiency.

show abstract

Efficient and Interpretable Deep Blind Image Deblurring Via Algorithm Unrolling

Cited by 163 publications

References 69 publications

HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment

HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment

Graph Signal Denoising Via Unrolling Networks

Model-Inspired Deep Learning for Light-Field Microscopy with Application to Neuron Localization

Contact Info

Product

Resources

About