Efficient Image Super-Resolution Using Pixel Attention

Zhao, Hengyuan; Kong, Xiangtao; He, Jingwen; Qiao, Yu; Dong, Chao

doi:10.1007/978-3-030-67070-2_3

Cited by 273 publications

(151 citation statements)

References 36 publications

Supporting

Mentioning

116

Contrasting

Order By: Relevance

“…The BI degradation model has been widely used to obtain LR images in the image SR tasks. In order to demonstrate the effectiveness of the RTAN, we compared it with 16 state-of-the-art CNN-based SR methods, including SRMDNF [7], NLRN [32], EDSR [17], DBPN [73], NDRCN [35], ACNet [38], FALSR-A [37], OISR-RK2-s [34], MCAN [47], A 2 F-SD [48], A2N-M [63], DeFiAN S [61], IMDN [33], SMSR [36], PAN [59], MGAN [62], RNAN [55].…”

Section: Results With Bicubic (Bi) Degradation Modelmentioning

confidence: 99%

“…In order to demonstrate the powerful reconstruction ability of the proposed method with BD degradation model, we compare the RTAN with 14 state-of-the-art CNN-based models, i.e., SPMSR [4], SRCNN [5], FSRCNN [74], VDSR [12], SRMD [7], EDSR [17], RDN [16], IRCNN [75], SRFBN [76], RCAN [6], A 2 F-SD [48], IMDN [33], DeFiAN S [61], PAN [59], and MGAN [62].…”

Section: Results With Blur-downscale (Bd) Degradation Modelmentioning

confidence: 99%

“…Muqeet et al [58] proposed the efficient MAFFSRN by modifying the ESA with the dilated convolutions to refine the features. Zhao et al [59] introduced the pixel attention mechanism to SR and designed the effective PAN. Mei et al [60] designed the pyramid attention network (PANet) to capture multi-scale feature correspondences.…”

Section: Attention Mechanismmentioning

confidence: 99%

See 2 more Smart Citations

Residual Triplet Attention Network for Single-Image Super-Resolution

Huang

Wang

et al. 2021

Electronics

View full text Add to dashboard Cite

Single-image super-resolution (SISR) techniques have been developed rapidly with the remarkable progress of convolutional neural networks (CNNs). The previous CNNs-based SISR techniques mainly focus on the network design while ignoring the interactions and interdependencies between different dimensions of the features in the middle layers, consequently hindering the powerful learning ability of CNNs. In order to address this problem effectively, a residual triplet attention network (RTAN) for efficient interactions of the feature information is proposed. Specifically, we develop an innovative multiple-nested residual group (MNRG) structure to improve the learning ability for extracting the high-frequency information and train a deeper and more stable network. Furthermore, we present a novel lightweight residual triplet attention module (RTAM) to obtain the cross-dimensional attention weights of the features. The RTAM combines two cross-dimensional interaction blocks (CDIBs) and one spatial attention block (SAB) base on the residual module. Therefore, the RTAM is not only capable of capturing the cross-dimensional interactions and interdependencies of the features, but also utilizing the spatial information of the features. The simulation results and analysis show the superiority of the proposed RTAN over the state-of-the-art SISR networks in terms of both evaluation metrics and visual results.

show abstract

Section: Results With Bicubic (Bi) Degradation Modelmentioning

confidence: 99%

Section: Results With Blur-downscale (Bd) Degradation Modelmentioning

confidence: 99%

See 1 more Smart Citation

Residual Triplet Attention Network for Single-Image Super-Resolution

Huang

Wang

et al. 2021

Electronics

View full text Add to dashboard Cite

show abstract

“…In the earlier densely connected structures, such as DenseNet [47] and SRDenseNet [48], the output results of each level of convolution or modules are directly added or concatenated as the input of the next level. In order to better integrate the output results of all levels of modules to assist in the reconstruction of the final features, when the bypass connection is established, inspired by PAN [49], we propose the Multi-Level Feature Fusion Block (MLFFB) based on the spatial attention mechanism, which can generate more detailed features. The network structure of the MLFFB is illustrated in Figure 4.…”

Section: Multi-level Feature Fusion Blockmentioning

confidence: 99%

Multi-Scale Feature Mapping Network for Hyperspectral Image Super-Resolution

et al. 2021

View full text Add to dashboard Cite

Hyperspectral Image (HSI) can continuously cover tens or even hundreds of spectral segments for each spatial pixel. Limited by the cost and commercialization requirements of remote sensing satellites, HSIs often lose a lot of information due to insufficient image spatial resolution. For the high-dimensional nature of HSIs and the correlation between the spectra, the existing Super-Resolution (SR) methods for HSIs have the problems of excessive parameter amount and insufficient information complementarity between the spectra. This paper proposes a Multi-Scale Feature Mapping Network (MSFMNet) based on the cascaded residual learning to adaptively learn the prior information of HSIs. MSFMNet simplifies each part of the network into a few simple yet effective network modules. To learn the spatial-spectral characteristics among different spectral segments, a multi-scale feature generation and fusion Multi-Scale Feature Mapping Block (MSFMB) based on wavelet transform and spatial attention mechanism is designed in MSFMNet to learn the spectral features between different spectral segments. To effectively improve the multiplexing rate of multi-level spectral features, a Multi-Level Feature Fusion Block (MLFFB) is designed to fuse the multi-level spectral features. In the image reconstruction stage, an optimized sub-pixel convolution module is used for the up-sampling of different spectral segments. Through a large number of verifications on the three general hyperspectral datasets, the superiority of this method compared with the existing hyperspectral SR methods is proved. In subjective and objective experiments, its experimental performance is better than its competitors.

show abstract

“…As shown in Fig. 2 (c), our CSRM is a multi-branch structure with pixel-attention modules [20] in each branch and contains several channel attention modules between two branches. We employ a 1×1 convolution layer at the beginning to reduce feature dimension by half.…”

Section: (B)mentioning

confidence: 99%

Residual Attention Block Search for Lightweight Image Super-Resolution

Liao

Zhao

Shen

et al. 2021

2021 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

Recently, lightweight neural networks with different manual designs have presented a promising performance in single image super-resolution (SR). However, these designs rely on too much expert experience. To address this issue, we focus on searching a lightweight block for efficient and accurate image SR. Due to the frequent use of various residual blocks and attention mechanisms in SR methods, we propose the residual attention search block (RASB) which combines an operation search block (OSB) with an attention search block (ASB). The former is used to explore the suitable operation at the proper position, and the latter is applied to discover the optimal connection of various attention mechanisms. Moreover, we build the modified residual attention network (MRAN) with stacked found blocks and a refinement module. Extensive experiments demonstrate that our MRAN achieves a better trade-off against the state-of-the-art methods in terms of accuracy and model complexity.

show abstract

Efficient Image Super-Resolution Using Pixel Attention

Cited by 273 publications

References 36 publications

Residual Triplet Attention Network for Single-Image Super-Resolution

Residual Triplet Attention Network for Single-Image Super-Resolution

Multi-Scale Feature Mapping Network for Hyperspectral Image Super-Resolution

Residual Attention Block Search for Lightweight Image Super-Resolution

Contact Info

Product

Resources

About