Enhanced Deep Residual Networks for Single Image Super-Resolution

@article{Lim2017EnhancedDR,
  title={Enhanced Deep Residual Networks for Single Image Super-Resolution},
  author={Bee Lim and Sanghyun Son and Heewon Kim and Seungjun Nah and Kyoung Mu Lee},
  journal={2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
  year={2017},
  pages={1132-1140}
}
  • Bee Lim, Sanghyun Son, Kyoung Mu Lee
  • Published 10 July 2017
  • Computer Science
  • 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Recent research on super-resolution has progressed with the development of deep convolutional neural networks (DCNN. [] Key Method The performance is further improved by expanding the model size while we stabilize the training procedure. We also propose a new multi-scale deep super-resolution system (MDSR) and training method, which can reconstruct high-resolution images of different upscaling factors in a single model. The proposed methods show superior performance over the state-of-the-art methods on…

Figures and Tables from this paper

Adaptive deep residual network for single image super-resolution

TLDR
A single image super-resolution model based on Adaptive Deep Residual named as ADR-SR, which uses the Input Output Same Size (IOSS) structure, and releases the dependence of upsampling layers compared with the existing SR methods.

Progressive residual networks for image super-resolution

TLDR
A novel Progressive Residual Network (PRNet) is proposed to integrate hierarchical and scale features for single image SR, which works well for both small and large scaling factors.

Multi-Branch Deep Residual Network for Single Image Super-Resolution

TLDR
The proposed method mainly uses three steps of feature extraction, mapping, and reconstruction to complete the super-resolution reconstruction and shows superior performance than other state-of-the-art super- Resolution methods on benchmark datasets.

Gradual deep residual network for super-resolution

TLDR
A gradual deep residual network for super-resolution (GDSR), which consists of multiple reconstruction network with 2 scale factor (2X reconstruction network), which outperforms the state-of-the-art methods in terms of quantitative evaluation, visual evaluation, and execution time evaluation.

Multilevel Residual Learning for Single Image Super Resolution

TLDR
A novel multilevel residual learning pattern for SISR is proposed that shows a stable performance improvement over the compared structures on several benchmark datasets with equal model parameters and empirically shows that simply increasing the number of building blocks will not obtain the expected improvements of performance.

Multi-scale Residual Network for Image Super-Resolution

TLDR
A novel multi-scale residual network (MSRN) to fully exploit the image features, which outperform most of the state-of-the-art methods.

Deep Residual Network with Enhanced Upscaling Module for Super-Resolution

  • Jun-Hyuk KimJong-Seok Lee
  • Computer Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • 2018
TLDR
This work argues that the single image super-resolution method has room for improvement, and proposes an enhanced upscaling module (EUM), which achieves improvement by utilizing nonlinear operations and skip connections.

Time accelerated image super-resolution using shallow residual feature representative network

TLDR
This work has developed an innovative shallow residual feature representative network (SRFRN) that uses a bicubic interpolated low-resolution image as input and residual representative units (RFR) which include serially stacked residual non-linear convolutions.

Single Image Super-Resolution Using Dual-Branch Convolutional Neural Network

TLDR
A novel skip connection is proposed to learn residual mapping and ease the training of the proposed DBCN, achieving comparable results in terms of accuracy and visual quality compared with several state-of-the-art methods.
...

References

SHOWING 1-10 OF 35 REFERENCES

Accelerating the Super-Resolution Convolutional Neural Network

TLDR
This paper aims at accelerating the current SRCNN, and proposes a compact hourglass-shape CNN structure for faster and better SR, and presents the parameter settings that can achieve real-time performance on a generic CPU while still maintaining good performance.

Learning a Deep Convolutional Network for Image Super-Resolution

TLDR
This work proposes a deep learning method for single image super-resolution (SR) that directly learns an end-to-end mapping between the low/high-resolution images and shows that traditional sparse-coding-based SR methods can also be viewed as a deep convolutional network.

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

TLDR
This paper presents the first convolutional neural network capable of real-time SR of 1080p videos on a single K2 GPU and introduces an efficient sub-pixel convolution layer which learns an array of upscaling filters to upscale the final LR feature maps into the HR output.

Accurate Image Super-Resolution Using Very Deep Convolutional Networks

TLDR
This work presents a highly accurate single-image superresolution (SR) method using a very deep convolutional network inspired by VGG-net used for ImageNet classification and uses extremely high learning rates enabled by adjustable gradient clipping.

Deeply-Recursive Convolutional Network for Image Super-Resolution

TLDR
This work proposes an image super-resolution method (SR) using a deeply-recursive convolutional network (DRCN) with two extensions: recursive-supervision and skip-connection, which outperforms previous methods by a large margin.

NTIRE 2017 Challenge on Single Image Super-Resolution: Methods and Results

TLDR
This paper reviews the first challenge on single image super-resolution (restoration of rich details in an low resolution image) with focus on proposed solutions and results and gauges the state-of-the-art in single imagesuper-resolution.

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

TLDR
SRGAN, a generative adversarial network (GAN) for image super-resolution (SR), is presented, to its knowledge, the first framework capable of inferring photo-realistic natural images for 4x upscaling factors and a perceptual loss function which consists of an adversarial loss and a content loss.

Learning Super-Resolution Jointly From External and Internal Examples

TLDR
Joint SR is proposed to adaptively combine the advantages of both external and internal SR methods, and defines two loss functions using sparse coding-based external examples, and epitomic matching based on internal examples, as well as a corresponding adaptive weight to automatically balance their contributions according to their reconstruction errors.

Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections

TLDR
This paper proposes to symmetrically link convolutional and de-convolutional layers with skip-layer connections, with which the training converges much faster and attains a higher-quality local optimum, making training deep networks easier and achieving restoration performance gains consequently.

Fast Direct Super-Resolution by Simple Functions

TLDR
This paper proposes to split the feature space into numerous subspaces and collect exemplars to learn priors for each subspace, thereby creating effective mapping functions and facilitating both feasibility of using simple functions for super-resolution, and efficiency of generating high-resolution results.