A Lightweight Dense Connected Approach with Attention on Single Image Super-Resolution

In recent years, neural networks for single image super-resolution (SISR) have applied more profound and deeper network structures to extract extra image details, which brings difficulties in model training. To deal with deep model training problems, researchers utilize dense skip connections to promote the model’s feature representation ability by reusing deep features of different receptive fields. Benefiting from the dense connection block, SRDensenet has achieved excellent performance in SISR. Despite the fact that the dense connected structure can provide rich information, it will also introduce redundant and useless information. To tackle this problem, in this paper, we propose a Lightweight Dense Connected Approach with Attention for Single Image Super-Resolution (LDCASR), which employs the attention mechanism to extract useful information in channel dimension. Particularly, we propose the recursive dense group (RDG), consisting of Dense Attention Blocks (DABs), which can obtain more significant representations by extracting deep features with the aid of both dense connections and the attention module, making our whole network attach importance to learning more advanced feature information. Additionally, we introduce the group convolution in DABs, which can reduce the number of parameters to 0.6 M. Extensive experiments on benchmark datasets demonstrate the superiority of our proposed method over five chosen SISR methods.

Download Full-text

Lightweight Single Image Super-resolution with Dense Connection Distillation Network

ACM Transactions on Multimedia Computing Communications and Applications ◽

10.1145/3414838 ◽

2021 ◽

Vol 17 (1s) ◽

pp. 1-17

Author(s):

Yanchun Li ◽

Jianglian Cao ◽

Zhetao Li ◽

Sangyoon Oh ◽

Nobuyoshi Komuro

Keyword(s):

Feature Fusion ◽

Super Resolution ◽

Single Image ◽

Feature Maps ◽

Deep Networks ◽

Benchmark Datasets ◽

Rich Information ◽

Image Super Resolution ◽

Single Image Super Resolution ◽

Training Efficiency

Single image super-resolution attempts to reconstruct a high-resolution (HR) image from its corresponding low-resolution (LR) image, which has been a research hotspot in computer vision and image processing for decades. To improve the accuracy of super-resolution images, many works adopt very deep networks to model the translation from LR to HR, resulting in memory and computation consumption. In this article, we design a lightweight dense connection distillation network by combining the feature fusion units and dense connection distillation blocks (DCDB) that include selective cascading and dense distillation components. The dense connections are used between and within the distillation block, which can provide rich information for image reconstruction by fusing shallow and deep features. In each DCDB, the dense distillation module concatenates the remaining feature maps of all previous layers to extract useful information, the selected features are then assessed by the proposed layer contrast-aware channel attention mechanism, and finally the cascade module aggregates the features. The distillation mechanism helps to reduce training parameters and improve training efficiency, and the layer contrast-aware channel attention further improves the performance of model. The quality and quantity experimental results on several benchmark datasets show the proposed method performs better tradeoff in term of accuracy and efficiency.

Download Full-text

Cascading and Residual Connected Network for Single Image Superresolution

Wireless Communications and Mobile Computing ◽

10.1155/2021/5579090 ◽

2021 ◽

Vol 2021 ◽

pp. 1-13

Author(s):

Kai Huang ◽

Wenhao Wang ◽

Cheng Pang ◽

Rushi Lan ◽

Ji Li ◽

...

Keyword(s):

Feature Learning ◽

Super Resolution ◽

Poor Performance ◽

Single Image ◽

Benchmark Datasets ◽

Image Super Resolution ◽

Model Training ◽

Single Image Super Resolution ◽

Residual Block ◽

Quantitative Indicators

Convolution neural networks facilitate the significant process of single image super-resolution (SISR). However, most of the existing CNN-based models suffer from numerous parameters and excessively deeper structures. Moreover, these models relying on in-depth features commonly ignore the hints of low-level features, resulting in poor performance. This paper demonstrates an intriguing network for SISR with cascading and residual connections (CASR), which alleviates these problems by extracting features in a small net called head module via the strategies based on the depthwise separable convolution and deformable convolution. Moreover, we also include a cascading residual block (CAS-Block) for the upsampling process, which benefits the gradient propagation and feature learning while easing the model training. Extensive experiments conducted on four benchmark datasets demonstrate that the proposed method is superior to the latest SISR methods in terms of quantitative indicators and realistic visual effects.

Download Full-text

Triple-Attention Mixed-Link Network for Single-Image Super-Resolution

Applied Sciences ◽

10.3390/app9152992 ◽

2019 ◽

Vol 9 (15) ◽

pp. 2992 ◽

Cited By ~ 1

Author(s):

Xi Cheng ◽

Xiang Li ◽

Jian Yang

Keyword(s):

Receptive Fields ◽

Super Resolution ◽

Feature Representation ◽

Single Image ◽

Deep Convolutional Neural Networks ◽

Flow Topology ◽

Image Super Resolution ◽

Representational Power ◽

Pass Through ◽

Single Image Super Resolution

Single-image super-resolution is of great importance as a low-level computer-vision task. Recent approaches with deep convolutional neural networks have achieved impressive performance. However, existing architectures have limitations due to the less sophisticated structure along with less strong representational power. In this work, to significantly enhance the feature representation, we proposed triple-attention mixed-link network (TAN), which consists of (1) three different aspects (i.e., kernel, spatial, and channel) of attention mechanisms and (2) fusion of both powerful residual and dense connections (i.e., mixed link). Specifically, the network with multi-kernel learns multi-hierarchical representations under different receptive fields. The features are recalibrated by the effective kernel and channel attention, which filters the information and enables the network to learn more powerful representations. The features finally pass through the spatial attention in the reconstruction network, which generates a fusion of local and global information, lets the network restore more details, and improves the reconstruction quality. The proposed network structure decreases 50% of the parameter growth rate compared with previous approaches. The three attention mechanisms provide 0.49 dB, 0.58 dB, and 0.32 dB performance gain when evaluating on Set5, Set14, and BSD100. Thanks to the diverse feature recalibrations and the advanced information flow topology, our proposed model is strong enough to perform against the state-of-the-art methods on the benchmark evaluations.

Download Full-text

Real-Time Environment Monitoring Using a Lightweight Image Super-Resolution Network

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph18115890 ◽

2021 ◽

Vol 18 (11) ◽

pp. 5890

Author(s):

Qiang Yu ◽

Feiqiang Liu ◽

Long Xiao ◽

Zitao Liu ◽

Xiaomin Yang

Keyword(s):

Deep Learning ◽

Real Time ◽

Super Resolution ◽

Model Complexity ◽

Practical Application ◽

Single Image ◽

Feature Maps ◽

Benchmark Datasets ◽

Image Super Resolution ◽

Single Image Super Resolution

Deep-learning (DL)-based methods are of growing importance in the field of single image super-resolution (SISR). The practical application of these DL-based models is a remaining problem due to the requirement of heavy computation and huge storage resources. The powerful feature maps of hidden layers in convolutional neural networks (CNN) help the model learn useful information. However, there exists redundancy among feature maps, which can be further exploited. To address these issues, this paper proposes a lightweight efficient feature generating network (EFGN) for SISR by constructing the efficient feature generating block (EFGB). Specifically, the EFGB can conduct plain operations on the original features to produce more feature maps with parameters slightly increasing. With the help of these extra feature maps, the network can extract more useful information from low resolution (LR) images to reconstruct the desired high resolution (HR) images. Experiments conducted on the benchmark datasets demonstrate that the proposed EFGN can outperform other deep-learning based methods in most cases and possess relatively lower model complexity. Additionally, the running time measurement indicates the feasibility of real-time monitoring.

Download Full-text

Improved differentiable neural architecture search for single image super-resolution

Peer-to-Peer Networking and Applications ◽

10.1007/s12083-020-01048-4 ◽

2021 ◽

Author(s):

Yu Weng ◽

Zehua Chen ◽

Tianbao Zhou

Keyword(s):

Super Resolution ◽

Poor Performance ◽

Machine Learning Algorithms ◽

Single Image ◽

Neural Architecture ◽

Deep Architecture ◽

Benchmark Datasets ◽

Image Super Resolution ◽

Classification Tasks ◽

Single Image Super Resolution

AbstractDeep learning has shown prominent superiority over other machine learning algorithms in Single Image Super-Resolution (SISR). In order to reduce the efforts and resources cost on manually designing deep architecture, we use differentiable neural architecture search (DARTS) on SISR. Since neural architecture search was originally used for classification tasks, our experiments show that direct usage of DARTS on super-resolutions tasks will give rise to many skip connections in the search architecture, which results in the poor performance of final architecture. Thus, it is necessary for DARTS to have made some improvements for the application in the field of SISR. According to characteristics of SISR, we remove redundant operations and redesign some operations in the cell to achieve an improved DARTS. Then we use the improved DARTS to search convolution cells as a nonlinear mapping part of super-resolution network. The new super-resolution architecture shows its effectiveness on benchmark datasets and DIV2K dataset.

Download Full-text

Single image super resolution using dictionary learning and sparse coding with multi-scale and multi-directional Gabor feature representation

Information Sciences ◽

10.1016/j.ins.2019.10.040 ◽

2020 ◽

Vol 512 ◽

pp. 1264-1278 ◽

Cited By ~ 3

Author(s):

Selen Ayas ◽

Murat Ekinci

Keyword(s):

Dictionary Learning ◽

Sparse Coding ◽

Super Resolution ◽

Feature Representation ◽

Single Image ◽

Multi Scale ◽

Gabor Feature ◽

Image Super Resolution ◽

Single Image Super Resolution

Download Full-text

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

Algorithms ◽

10.3390/a11100144 ◽

2018 ◽

Vol 11 (10) ◽

pp. 144 ◽

Cited By ~ 2

Author(s):

Peng Liu ◽

Ying Hong ◽

Yan Liu

Keyword(s):

Super Resolution ◽

Superior Performance ◽

Evaluation Function ◽

Single Image ◽

Residual Network ◽

Perceptual Evaluation ◽

Benchmark Datasets ◽

Image Super Resolution ◽

Single Image Super Resolution

Recently, algorithms based on the deep neural networks and residual networks have been applied for super-resolution and exhibited excellent performance. In this paper, a multi-branch deep residual network for single image super-resolution (MRSR) is proposed. In the network, we adopt a multi-branch network framework and further optimize the structure of residual network. By using residual blocks and filters reasonably, the model size is greatly expanded while the stable training is also guaranteed. Besides, a perceptual evaluation function, which contains three parts of loss, is proposed. The experiment results show that the evaluation function provides great support for the quality of reconstruction and the competitive performance. 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.

Download Full-text

Deeply Recursive Low- and High-Frequency Fusing Networks for Single Image Super-Resolution

Sensors ◽

10.3390/s20247268 ◽

2020 ◽

Vol 20 (24) ◽

pp. 7268

Author(s):

Cheng Yang ◽

Guanming Lu

Keyword(s):

High Frequency ◽

Super Resolution ◽

Single Image ◽

Feature Representations ◽

Prediction Time ◽

Reconstruction Quality ◽

Benchmark Datasets ◽

Image Super Resolution ◽

Single Image Super Resolution ◽

Memory Resources

With the development of researches on single image super-resolution (SISR) based on convolutional neural networks (CNN), the quality of recovered images has been remarkably promoted. Since then, many deep learning-based models have been proposed, which have outperformed the traditional SISR algorithms. According to the results of extensive experiments, the feature representations of the model can be enhanced by increasing the depth and width of the network, which can ultimately improve the image reconstruction quality. However, a larger network generally consumes more computational and memory resources, making it difficult to train the network and increasing the prediction time. In view of the above problems, a novel deeply-recursive low- and high-frequency fusing network (DRFFN) for SISR tasks is proposed in this paper, which adopts the structure of parallel branches to extract the low- and high-frequency information of the image, respectively. The different complexities of the branches can reflect the frequency characteristic of the diverse image information. Moreover, an effective channel-wise attention mechanism based on variance (VCA) is designed to make the information distribution of each feature map more reasonably with different variances. Owing to model structure (i.e., cascading recursive learning of recursive units), DRFFN and DRFFN-L are very compact, where the weights are shared by all convolutional recursions. Comprehensive benchmark evaluations in standard benchmark datasets well demonstrate that DRFFN outperforms the most existing models and has achieved competitive, quantitative, and visual results.

Download Full-text