Fast Convolutional Sparse Coding

By Hilton Bristow, Anders P. Eriksson, Simon Lucey. 2013 IEEE Conference on Computer Vision and Pattern Recognition, 2013

364 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/6c0eb85c99044d19cc59b2cf46b20eadb302ce0e

How researchers see this paper

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

In Computational Mechanics

1. Lei Zhang
2. Wangmeng Zuo
3. Xiangchu Feng
4. Deyu Meng
5. Shuhang Gu
6. Yaniv Romano
7. Vardan Papyan
8. Yu Liu

In Computer Vision and Pattern Recognition

1. Lei Zhang
2. Luc Van Gool
3. Wangmeng Zuo
4. Ming-Hsuan Yang
5. Chen Change Loy
6. Dacheng Tao
7. Andrea Vedaldi
8. Radu Timofte

In Signal Processing

1. Lei Zhang
2. Xiangchu Feng
3. Zhijin Qin
4. Vardan Papyan
5. Rabab Kreidieh Ward
6. Geoffrey Y. Li
7. Yaniv Romano
8. Yu Liu

Citations in the corpus

• Semantic Communications: Principles and Challenges (2021, background): “Later on, sparse coding [50] was introduced to represent image patches as a combination of overcomplete basis elements, also known as a codebook.”
• Efficient Algorithms for Convolutional Sparse Representations (2016, methodology): “Most of these methods solve the problem in the spatial domain, but [21], [35], [37], [38], [43], [44] exploit the convolution theorem to solve at least part of the problem in the Discrete Fourier Transform (DFT) domain.”
• Convolutional Sparse Coding for Image Super-Resolution (2015, methodology): “[1] proposed a fast CSC algorithm by considering the property of block circulant with circulant block (BCCB) matrix in the Fourier doThe Training Phase”
• Convolutional Neural Networks Analyzed via Convolutional Sparse Coding (2016, methodology): “However, in recent years an alternative to this patch-based processing has emerged in the form of the Convolutional Sparse Coding (CSC) model (Bristow et al., 2013; Kong and Fowlkes, 2014; Wohlberg, 2014; Gu et al., 2015; Heide et al., 2015; Papyan et al., 2016a,b).”
• Deep Image Prior (2017, methodology): “Also in this group are methods based on fitting dictionaries to the patches of the corrupted image [40,62] as well as methods based on convolutional sparse coding [21, 7].”
• Image Fusion With Convolutional Sparse Representation (2016, methodology): “In the past few years, several efficient algorithms [22], [28], [29] within the alternating direction method of multipliers [30] framework have been proposed to solve this problem in the frequency”
• Compressed Sensing MRI Reconstruction Using a Generative Adversarial Network With a Cyclic Loss (2017, methodology): “2D and 3D CSC have been successfully adopted in dynamic CS-MRI [20], [21], and efficient numerical algorithms, such as the alternating direction method of multipliers, are proposed to further accelerate CSC computation in CS-MRI [22], [23] Deep learning-based methods: Deep learning-based CSMRI is ai”
• Hierarchical Convolutional Deep Learning in Computer Vision (2013, methodology): “Although some speed-up might be gained by using FFTs in place of spatial convolutions [14], particularly if the filter size NFrows by NFcols is large, this can introduce boundary effects in the feature maps – therefore solving in the spatial domain was used throughout this work.”