Second-Order Attention Network for Single Image Super-Resolution

Dai, Tao; Cai, Jianrui; Zhang, Yongbing; Xia, Shu-Tao; Zhang, Lei

doi:10.1109/cvpr.2019.01132

Cited by 1,436 publications

(938 citation statements)

References 25 publications

Supporting

Mentioning

827

Contrasting

Unclassified

Order By: Relevance

“…Attention or non-local modeling is one of the choices to globally capture the feature response across the whole image. A lot of related works [31,7,26,27,15,5] have been proposed for computing vision successfully. There are several advantages of using attention operations: 1) It can directly compute the correlation between patterns across the image regardless of their distances; 2) It can efficiently reduce the number of kernels and depth of the network to achieve comparable or even better performance and 3) Finally, it is also easy to be embedded into any structure for operations.…”

Section: Introductionmentioning

confidence: 99%

Image Super-Resolution via Attention Based Back Projection Networks

Liu

Wang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

Deep learning based image Super-Resolution (SR) has shown rapid development due to its ability of big data digestion. Generally, deeper and wider networks can extract richer feature maps and generate SR images with remarkable quality. However, the more complex network we have, the more time consumption is required for practical applications. It is important to have a simplified network for efficient image SR. In this paper, we propose an Attention based Back Projection Network (ABPN) for image superresolution. Similar to some recent works, we believe that the back projection mechanism can be further developed for SR. Enhanced back projection blocks are suggested to iteratively update low-and high-resolution feature residues. Inspired by recent studies on attention models, we propose a Spatial Attention Block (SAB) to learn the cross-correlation across features at different layers. Based on the assumption that a good SR image should be close to the original LR image after down-sampling. We propose a Refined Back Projection Block (RBPB) for final reconstruction. Extensive experiments on some public and AIM2019 Image Super-Resolution Challenge [4] datasets show that the proposed ABPN can provide state-of-the-art or even better performance in both quantitative and qualitative measurements.

show abstract

Section: Introductionmentioning

confidence: 99%

Image Super-Resolution via Attention Based Back Projection Networks

Liu

Wang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

show abstract

“…In order to verify the validity of the model , we compare the performance on five standard benchmark datasets: Set5 [1], Set14 [27], B100 [17], Urban100 [8], and manga109 [18]. In terms of PSNR, SSIM and visual effects, We compare our models with the state-of-theart methods including Bicubic, SRCNN [5], VDSR [10], LapSRN [12], MemNet [22], EDSR [16], RDN [32], RCAN [31], SAN [4]. We also adopt self-ensemble strategy [16] to further improve our ADCSR and denote the selfensembled ADCSR as ADCSR+.…”

Section: Results With Bicubic Degradationmentioning

confidence: 99%

“…As can be seen from the table, the PSNR and SSIM of the algorithm in ×2, ×3, ×4 exceed the current state of the art. Figure 6 show the Qualitative comparison of our models with Bicubic, SRCNN [5], VDSR [10], LapSRN [12], MSLapSRN [13], EDSR [16], RCAN [31], and SAN [4] . The images of SRCNN, EDSR, and RCAN are derived from the author's open-source model and code.…”

Section: Results With Bicubic Degradationmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Densely Connected Single Image Super-Resolution

Xie

Yang

Jia³

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

For a better performance in single image superresolution(SISR), we present an image super-resolution algorithm based on adaptive dense connection (ADCSR). The algorithm is divided into two parts: BODY and SKIP. BODY improves the utilization of convolution features through adaptive dense connections. Also, we develop an adaptive sub-pixel reconstruction layer (AFSL) to reconstruct the features of the BODY output. We pre-trained SKIP to make BODY focus on high-frequency feature learning. The comparison of PSNR, SSIM, and visual effects verify the superiority of our method to the state-of-the-art algorithms.

show abstract

“…Guo et al [32] proposed a DCT-DSR network to address the super-resolution problem in an image transform domain. To further enhance the quality of the image, Dai et al [33] and Li et al [35] proposed a secondorder attention mechanism and a feedback mechanism to perform super-resolution, respectively.…”

Section: Image Super-resolutionmentioning

confidence: 99%

FCSR-GAN: Joint Face Completion and Super-Resolution via Multi-Task Learning

Cai

Han

Shan

et al. 2020

IEEE Trans. Biom. Behav. Identity Sci.

View full text Add to dashboard Cite

Combined variations containing low-resolution and occlusion often present in face images in the wild, e.g., under the scenario of video surveillance. While most of the existing face image recovery approaches can handle only one type of variation per model, in this work, we propose a deep generative adversarial network (FCSR-GAN) for performing joint face completion and face super-resolution via multi-task learning. The generator of FCSR-GAN aims to recover a high-resolution face image without occlusion given an input low-resolution face image with occlusion. The discriminator of FCSR-GAN uses a set of carefully designed losses (an adversarial loss, a perceptual loss, a pixel loss, a smooth loss, a style loss, and a face prior loss) to assure the high quality of the recovered high-resolution face images without occlusion. The whole network of FCSR-GAN can be trained end-to-end using our two-stage training strategy. Experimental results on the public-domain CelebA and Helen databases show that the proposed approach outperforms the state-of-the-art methods in jointly performing face super-resolution (up to 8 ×) and face completion, and shows good generalization ability in cross-database testing. Our FCSR-GAN is also useful for improving face identification performance when there are low-resolution and occlusion in face images. The code of FCSR-GAN is available at: https://github.com/swordcheng/FCSR-GAN.

show abstract

Second-Order Attention Network for Single Image Super-Resolution

Cited by 1,436 publications

References 25 publications

Image Super-Resolution via Attention Based Back Projection Networks

Image Super-Resolution via Attention Based Back Projection Networks

Adaptive Densely Connected Single Image Super-Resolution

FCSR-GAN: Joint Face Completion and Super-Resolution via Multi-Task Learning

Contact Info

Product

Resources

About