Single Image Super-Resolution With Non-Local Means and Steering Kernel Regression

By Kaibing Zhang, Xinbo Gao, Dacheng Tao, Xuelong Li. IEEE Transactions on Image Processing, 2012

574 citations recorded by the source corpus.

Image super-resolution (SR) reconstruction is essentially an ill-posed problem, so it is important to design an effective prior. For this purpose, we propose a novel image SR method by learning both non-local and local regularization priors from a given low-resolution image. The non-local prior takes advantage of the redundancy of similar patches in natural images, while the local prior assumes that a target pixel can be estimated by a weighted average of its neighbors. Based on the above considerations, we utilize the non-local means filter to learn a non-local prior and the steering kernel regression to learn a local prior. By assembling the two complementary regularization terms, we propose a maximum a posteriori probability framework for SR recovery. Thorough experimental results suggest that the proposed SR method can reconstruct higher quality results both quantitatively and perceptually.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/8bbab35062c179a85bfc1822c8553050ac76de48

How researchers see this paper

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

In Computer Vision and Pattern Recognition

1. Lei Zhang
2. Luc Van Gool
3. Radu Timofte
4. Chao Dong
5. Yun Raymond Fu
6. Yulun Zhang
7. Kai Zhang
8. Gangshan Wu

In Media Technology

1. Lei Zhang
2. Chao Dong
3. Yulun Zhang
4. Radu Timofte
5. Yun Raymond Fu
6. Jie Tang
7. Jie Liu
8. Gangshan Wu

Citations in the corpus

• Residual Feature Aggregation Network for Image Super-Resolution (2020, methodology): “Many image SR methods have been proposed to tackle this inverse problem, including early interpolationbased [37], reconstruction-based [34], and recent learning based methods [27, 28, 22, 4, 12, 13, 36, 3].”
• Residual Dense Network for Image Super-Resolution (2018, methodology): “To tackle this inverse problem, plenty of image SR algorithms have been proposed, including interpolationbased [36], reconstruction-based [34], and learning-based methods [27, 28, 19, 2, 20, 8, 10, 30].”
• Second-Order Attention Network for Single Image Super-Resolution (2019, methodology): “During the past decade, a plenty of image SISR meth-ods have been proposed in the computer vision community, including interpolation-based [37], model-based [34], and CNN-based methods [2, 29, 14, 13, 29, 17, 30, 39, 38].”
• Feedback Network for Image Super-Resolution (2019, methodology): “To address this problem, numerous image SR methods have been proposed, including interpolation-based methods[45], reconstruction-based methods[42], and learning-based methods [33, 26, 34, 15, 29, 6, 18].”
• End-to-End Image Super-Resolution via Deep and Shallow Convolutional Networks (2016, unknown): “based [29]–[31], and learning-based methods [6], [9], [13].”
• Deep Networks for Image Super-Resolution with Sparse Prior (2015, unknown): “We also compare CSCN with other sparse coding extensions [22, 10, 38], and consider the blurring effect introduced in downscaling.”
• Residual Dense Network for Image Restoration (2018, background): “Recently, deep convolutional neural network (CNN) has been widely investigated and achieves promising performance in various image restoration tasks, such as image super-resolution (SR) [10], [11], [12], [13], [14],”