RAISR: Rapid and Accurate Image Super Resolution

@article{Romano2017RAISRRA,
  title={RAISR: Rapid and Accurate Image Super Resolution},
  author={Yaniv Romano and John R. Isidoro and Peyman Milanfar},
  journal={IEEE Transactions on Computational Imaging},
  year={2017},
  volume={3},
  pages={110-125}
}
Given an image, we wish to produce an image of larger size with significantly more pixels and higher image quality. This is generally known as the single image super-resolution problem. The idea is that with sufficient training data (corresponding pairs of low and high resolution images) we can learn set of filters (i.e., a mapping) that when applied to given image that is not in the training set, will produce a higher resolution version of it, where the learning is preferably low complexity… 
EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis
TLDR
This work proposes a novel application of automated texture synthesis in combination with a perceptual loss focusing on creating realistic textures rather than optimizing for a pixelaccurate reproduction of ground truth images during training to achieve a significant boost in image quality at high magnification ratios.
Pixel Recursive Super Resolution
TLDR
This work proposes a new probabilistic deep network architecture, a pixel recursive super resolution model, that is an extension of PixelCNNs to address the problem of artificially enlarging a low resolution photograph to recover a plausible high resolution version.
Image super resolution by dilated dense progressive network
TLDR
An effective model based on dilated dense network operations to accelerate deep networks for image SR, which support the exponential growth of the receptive field parallel by increasing the filter size is proposed.
Single image resolution enhancement by efficient dilated densely connected residual network
TLDR
An effective model based on progressive dilated densely connected and a novel activation function, which is appropriate for image SR problems is presented and achieves higher SR performance as compared to state-of-the-art results.
SINGLE IMAGE SUPER-RESOLUTION WITH LIMITED NUMBER OF FILTERS
TLDR
A single image super-resolution with limited number of filters based on RAISR, where only 18 filters can make high resolution images by using simple geometric conversion and rotation conversion while keeping the accuracy and runtime of RAisR.
Structure-Preserving Image Super-Resolution via Contextualized Multitask Learning
TLDR
This paper develops a contextualized multitask learning framework that first extracts convolutional features from the input LR image and applies one deconvolutional module to interpolate the LR feature maps in a content-adaptive way and demonstrates the effectiveness of the method on achieving both higher restoration quality and computational efficiency compared with several state-of-the-art SR approaches.
Super-Resolution based on Image-Adapted CNN Denoisers: Incorporating Generalization of Training Data and Internal Learning in Test Time
TLDR
This work applies a recent P&P strategy to SR, and shows how it may become image-adaptive in test time, and outperforms the above two strategies on popular datasets and gives better results than other state-of-the-art methods on real images.
Degradation Model Learning for Real-World Single Image Super-Resolution
TLDR
This work proposes to learn a realistic degradation model from the existing real-world datasets, and uses the learned degradation model to synthesize realistic HR-LR image pairs, and simultaneously learns a weight prediction network to predict the pixel-wise spatially variant degradation kernel as the weighted combination of the basis kernels.
Large Factor Image Super-Resolution With Cascaded Convolutional Neural Networks
TLDR
A cascaded super-resolution convolutional neural network (CSRCNN), which takes a single low-resolution image as an input and reconstructs high-resolution images in a progressive way, achieves superior results compared with those of other state-of-the-art methods, particularly with the 8× upsampling factor.
Supplementary: Practical Single-Image Super-Resolution Using Look-Up Table
With RAISR RAISR [9] works in two-stage that consists of a global and a local enhancement, with the latter part using a hash table. They compute image gradient and SVD for each patch of the input
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 59 REFERENCES
Image super-resolution as sparse representation of raw image patches
TLDR
It is shown that a small set of randomly chosen raw patches from training images of similar statistical nature to the input image generally serve as a good dictionary, in the sense that the computed representation is sparse and the recovered high-resolution image is competitive or even superior in quality to images produced by other SR methods.
Super-resolution image reconstruction: a technical overview
TLDR
The goal of this article is to introduce the concept of SR algorithms to readers who are unfamiliar with this area and to provide a review for experts to present the technical review of various existing SR methodologies which are often employed.
On Single Image Scale-Up Using Sparse-Representations
TLDR
This paper deals with the single image scale-up problem using sparse-representation modeling, and assumes a local Sparse-Land model on image patches, serving as regularization, to recover an original image from its blurred and down-scaled noisy version.
Global Image Denoising
TLDR
Experiments illustrate that the proposed global filter can effectively globalize any existing denoising filters to estimate each pixel using all pixels in the image, hence improving upon the best patch-based methods.
Super-resolution through neighbor embedding
  • Hong Chang, D. Yeung, Yimin Xiong
  • Computer Science
    Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.
  • 2004
TLDR
This paper proposes a novel method for solving single-image super-resolution problems, given a low-resolution image as input, and recovers its high-resolution counterpart using a set of training examples, inspired by recent manifold teaming methods.
A Statistical Prediction Model Based on Sparse Representations for Single Image Super-Resolution
TLDR
The suggested approach offers a desirable compromise between low computational complexity and reconstruction quality, when comparing it with state-of-the-art methods for single image super-resolution.
Image Super-Resolution Via Sparse Representation
TLDR
This paper presents a new approach to single-image superresolution, based upon sparse signal representation, which generates high-resolution images that are competitive or even superior in quality to images produced by other similar SR methods.
Image Super-Resolution Using Deep Convolutional Networks
We propose a deep learning method for single image super-resolution (SR). Our method directly learns an end-to-end mapping between the low/high-resolution images. The mapping is represented as a deep
Single Image Interpolation Via Adaptive Nonlocal Sparsity-Based Modeling
TLDR
This paper proposes a novel image interpolation method, which combines these two forces-nonlocal self-similarities and sparse representation modeling, and the proposed method is contrasted with competitive and related algorithms, and demonstrated to achieve state-of-the-art results.
A+: Adjusted Anchored Neighborhood Regression for Fast Super-Resolution
TLDR
This work proposes A+, an improved variant of Anchored Neighborhood Regression, which combines the best qualities of ANR and SF and builds on the features and anchored regressors from ANR but instead of learning the regressors on the dictionary it uses the full training material, similar to SF.
...
1
2
3
4
5
...