Single-Image Super-Resolution: A Benchmark

By Chih-Yuan Yang, Chao Ma, Ming-Hsuan Yang. 2014

608 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (2021, methodology): “As mentioned in the introduction, bicubic downsampling and traditional degradations underly existing DNNsbased SISR methods [22, 34, 41, 55], or some simple variants do [10, 11, 38, 47, 53, 54].”
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network (2016, methodology): “Over the last few decades substantial advances have been made in image SR [36, 53], with early methods based on interpolation, simple image features (e.”
Perceptual Losses for Real-Time Style Transfer and Super-Resolution (2016, background): “[28] provide an exhaustive evaluation of the prevailing techniques prior to the widespread adoption of convolutional neural networks.”
A Comprehensive guide to Bayesian Convolutional Neural Network with Variational Inference (2019, methodology): “licable to many areas like medical imaging [54], face recognition [21] and so on. There are many ways to do a single image super-resolution and detailed benchmarks of the methods are provided by Yang [62]. Following are the major ways to do a single image super-resolution: Prediction Models: These m”
The Perception-Distortion Tradeoff (2017, neutral): “PSNR and SSIM [22, 16, 6, 38, 51, 49].”
Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (2016, methodology): “An alternative family of methods are single image super-resolution (SISR) techniques [45].”
EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis (2016, neutral): “As shown in the supplementary, SSIM and IFC [46] which have been found to correlate better with human perception [57] also do not capture the perceptual quality of the results, so we provide alternative quantitative evaluations that agree better with human perception in Sec.”
Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution (2017, methodology): “In particular, our algorithm achieves higher IFC values, which has been shown to be correlated well with human perception of image super-resolution [35].”
Learning a No-Reference Quality Metric for Single-Image Super-Resolution (2016, methodology): “In our recent SR benchmark work [17], we conduct subject studies in a subset of generated SR images, and show that the IFC [18] metric performs well among full-reference measures.”
Physics-informed machine learning: case studies for weather and climate modelling (2021, background): “Super-resolution (SR) is the process of taking a low-resolution (LR) image and producing an enhanced image that approximates the true high-resolution (HR) version of it [116].”
NTIRE 2019 Challenge on Video Deblurring and Super-Resolution: Dataset and Study (2019, background): “There have been several studies about deblurring and superresolution which aimed to improve perceptual quality with adversarial and perceptual losses [23, 17, 18, 10, 39].”
Deep Plug-And-Play Super-Resolution for Arbitrary Blur Kernels (2019, background): “Empirical and theoretical studies have proven that an accurate degradation model is crucial for the success of SISR [20, 60].”
Toward Real-World Single Image Super-Resolution: A New Benchmark and a New Model (2019, methodology): “adation). Unfortunately, SISR models trained on such simulated datasets are hard to generalize to practical applications since the authentic degradations in real-world LR images are much more complex [54,26]. Fig.1shows the SISR results of a realworld image captured by a Sony a7II camera. We utilize”
Fast and accurate image upscaling with super-resolution forests (2015, result): “We report both PSNR (in dB) and the IFC score, which was shown to have higher correlation with the human perception compared to other metrics [36].”