ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Cornet: Composite-Regularized Neural Network for Convolutional Sparse Coding

Jawali, D and Pokala, PK and Seelamantula, CS (2020) Cornet: Composite-Regularized Neural Network for Convolutional Sparse Coding. In: Proceedings - International Conference on Image Processing, 25-28, September 2020, Virtual, Abu Dhabi; United Arab Emirates, pp. 818-822.

[img] PDF
Pro-Vol-2020C.pdf - Published Version
Restricted to Registered users only

Download (517kB) | Request a copy
Official URL: https://dx.doi.org/10.1109/ICIP40778.2020.9190923


Sparse recovery via composite regularization is an interesting approach proposed recently in the literature. One could design nonconvex regularizers through a convex combination of sparsitypromoting penalties with known proximal operators. We develop a new algorithm, namely, convolutional proximal-averaged thresholding algorithm (C-PATA) for composite-regularized convolutional sparse coding (CR-CSC) based on the recently proposed idea of proximal averaging. We develop an autoencoder structure based on the deep-unfolding of C-PATA iterations into neural network layers, which results in the composite-regularized neural network (CoRNet) architecture. The convolutional learned iterative soft-thresholding algorithm becomes a special case of CoRNet. We demonstrate the efficacy of CoRNet considering applications to image denoising and inpainting, and compare the performance with state-of-the-art techniques such as BM3D, convolutional LISTA, and fast and flexible convolutional sparse coding (FFCSC). © 2020 IEEE.

Item Type: Conference Paper
Publication: Proceedings - International Conference on Image Processing, ICIP
Publisher: IEEE Computer Society
Additional Information: cited By 0; Conference of 2020 IEEE International Conference on Image Processing, ICIP 2020 ; Conference Date: 25 September 2020 Through 28 September 2020; Conference Code:165772
Keywords: Composite structures; Convolution; Image coding; Image denoising; Iterative methods; Multilayer neural networks; Network coding; Network layers, Convex combinations; Regularized neural networks; Soft-thresholding algorithm; Sparse coding; Sparse recovery; State-of-the-art techniques; Structure-based; Thresholding algorithms, Convolutional neural networks
Department/Centre: Division of Electrical Sciences > Electrical Engineering
Date Deposited: 22 Jan 2021 10:42
Last Modified: 22 Jan 2021 10:42
URI: http://eprints.iisc.ac.in/id/eprint/67730

Actions (login required)

View Item View Item