NTIRE 2018 Challenge on Single Image Super-Resolution: Methods and Results

Timofte, Radu; Gu, Shuhang; Gool, Luc Van; Zhang, Lei; Yang, Ming–Hsuan

doi:10.1109/cvprw.2018.00130

Cited by 286 publications

(193 citation statements)

References 26 publications

Supporting

Mentioning

191

Contrasting

Unclassified

Order By: Relevance

“…We implement D-DBPN-L which is a dense connection of the L network to show how dense connection can improve the network's performance in all cases as shown in Table 5. On 4× enlargement, the dense network, D-DBPN- [27] and PIRM2018 [28], uses n 0 = 256, n R = 64, and t = 10 for the back-projection stages, and dense connection between projection units. In the reconstruction, we use conv ( Table 6.…”

Section: Comparison Of Each Dbpn Variantmentioning

confidence: 99%

See 1 more Smart Citation

Deep Back-Projection Networks for Super-Resolution

Haris

Shakhnarovich

Ukita

2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

Self Cite

1,328

940

View full text Add to dashboard Cite

Previous feed-forward architectures of recently proposed deep super-resolution networks learn the features of low-resolution inputs and the non-linear mapping from those to a high-resolution output. However, this approach does not fully address the mutual dependencies of low-and high-resolution images. We propose Deep Back-Projection Networks (DBPN), the winner of two image super-resolution challenges (NTIRE2018 and PIRM2018), that exploit iterative up-and down-sampling layers. These layers are formed as a unit providing an error feedback mechanism for projection errors. We construct mutually-connected up-and down-sampling units each of which represents different types of image degradation and high-resolution components. We also show that extending this idea to several variants applying the latest deep network trends, such as recurrent network, dense connection, and residual learning, to improve the performance. The experimental results yield superior results and in particular establishing new state-of-the-art results across multiple data sets, especially for large scaling factors such as 8×.

show abstract

Section: Comparison Of Each Dbpn Variantmentioning

confidence: 99%

“…Furthermore, DBPN has been proven by winning SISR challenges. On NTIRE2018 [27], DBPN is the 1 st winner on track 8× Bicubic downscaling. On PIRM2018 [28], DBPN got 1 st on Region 2, 3 rd on Region 1, and 5 th on Region 3.…”

Section: Introductionmentioning

confidence: 99%

Deep Back-Projection Networks for Super-Resolution

Haris

Shakhnarovich

Ukita

2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

Self Cite

1,328

940

View full text Add to dashboard Cite

show abstract

“…For instance, the bicubic downsampling D Bic (·) formulates the SR problem asX = S(D Bic (X)) and the Gaussian downsampling asX = S(D Gau (X)). For the more complicated degradation model imposed in [26], it isX = S(D Blur (D Bic (X)) + v), where D Blur (·) denotes a blurring operator and v denotes a certain kind of noise. Unlike the synthetic degradation models as mentioned above, it is difficult to derive an analytic expression for D RV (·).…”

Section: Problem Formulationmentioning

confidence: 99%

Camera Lens Super-Resolution

Chang

Xiong

Tian

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

151

View full text Add to dashboard Cite

Existing methods for single image super-resolution (SR) are typically evaluated with synthetic degradation models such as bicubic or Gaussian downsampling. In this paper, we investigate SR from the perspective of camera lenses, named as CameraSR, which aims to alleviate the intrinsic tradeoff between resolution (R) and field-of-view (V) in realistic imaging systems. Specifically, we view the R-V degradation as a latent model in the SR process and learn to reverse it with realistic low-and high-resolution image pairs. To obtain the paired images, we propose two novel data acquisition strategies for two representative imaging systems (i.e., DSLR and smartphone cameras), respectively. Based on the obtained City100 dataset, we quantitatively analyze the performance of commonly-used synthetic degradation models, and demonstrate the superiority of CameraSR as a practical solution to boost the performance of existing SR methods. Moreover, CameraSR can be readily generalized to different content and devices, which serves as an advanced digital zoom tool in realistic imaging systems.

show abstract

“…Since 2017, the NTIRE challenge [14] established itself as one of the major benchmarks for SISR. The challenge is based on the DIV2K dataset [15], a open dataset of hand-selected and diverse images obtained from the internet.…”

Section: Single Image Super-resolutionmentioning

confidence: 99%

Super-resolution of PROBA-V images using convolutional neural networks

et al. 2019

View full text Add to dashboard Cite

ESA's PROBA-V Earth observation satellite enables us to monitor our planet at a large scale, studying the interaction between vegetation and climate and provides guidance for important decisions on our common global future. However, the interval at which high resolution images are recorded spans over several days, in contrast to the availability of lower resolution images which is often daily. We collect an extensive dataset of both, high and low resolution images taken by PROBA-V instruments during monthly periods to investigate Multi Image Super-resolution, a technique to merge several low resolution images to one image of higher quality. We propose a convolutional neural network that is able to cope with changes in illumination, cloud coverage and landscape features which are challenges introduced by the fact that the different images are taken over successive satellite passages over the same region. Given a bicubic upscaling of low resolution images taken under optimal conditions, we find the Peak Signal to Noise Ratio of the reconstructed image of the network to be higher for a large majority of different scenes. This shows that applied machine learning has the potential to enhance large amounts of previously collected earth observation data during multiple satellite passes.

show abstract

NTIRE 2018 Challenge on Single Image Super-Resolution: Methods and Results

Cited by 286 publications

References 26 publications

Deep Back-Projection Networks for Super-Resolution

Deep Back-Projection Networks for Super-Resolution

Camera Lens Super-Resolution

Super-resolution of PROBA-V images using convolutional neural networks

Contact Info

Product

Resources

About