Image Super-Resolution Using Dense Skip Connections

Tong, Tong; Li, Gen; Liu, Xiejie; Gao, Qinquan

doi:10.1109/iccv.2017.514

Cited by 1,092 publications

(719 citation statements)

References 18 publications

Supporting

Mentioning

714

Contrasting

Unclassified

Order By: Relevance

“…The authors removed some unnecessary modules (e.g., Batch Normalization) of the SRResNet [16] to obtain better results. Based on EDSR, Zhang et al incorporated densely connected block [9,27] into residual block [7] to construct a residual dense network (RDN). Soon they exploited the residualin-residual architecture for the very deep model and introduced channel attention mechanism [8] to form the very deep residual attention networks (RCAN) [36].…”

Section: Introductionmentioning

confidence: 99%

Lightweight Image Super-Resolution with Information Multi-distillation Network

Hui

Gao

Yang

et al. 2019

Proceedings of the 27th ACM International Conference on Multimedia

588

426

View full text Add to dashboard Cite

In recent years, single image super-resolution (SISR) methods using deep convolution neural network (CNN) have achieved impressive results. Thanks to the powerful representation capabilities of the deep networks, numerous previous ways can learn the complex non-linear mapping between low-resolution (LR) image patches and their high-resolution (HR) versions. However, excessive convolutions will limit the application of super-resolution technology in low computing power devices. Besides, super-resolution of any arbitrary scale factor is a critical issue in practical applications, which has not been well solved in the previous approaches. To address these issues, we propose a lightweight information multi-distillation network (IMDN) by constructing the cascaded information multidistillation blocks (IMDB), which contains distillation and selective fusion parts. Specifically, the distillation module extracts hierarchical features step-by-step, and fusion module aggregates them according to the importance of candidate features, which is evaluated by the proposed contrast-aware channel attention mechanism. To process real images with any sizes, we develop an adaptive cropping strategy (ACS) to super-resolve block-wise image patches using the same well-trained model. Extensive experiments suggest that the proposed method performs favorably against the state-of-the-art SR algorithms in term of visual quality, memory footprint, and inference time. Code is available at https://github.com/Zheng222/IMDN.

show abstract

Section: Introductionmentioning

confidence: 99%

Lightweight Image Super-Resolution with Information Multi-distillation Network

Hui

Gao

Yang

et al. 2019

Proceedings of the 27th ACM International Conference on Multimedia

588

426

View full text Add to dashboard Cite

show abstract

“…To start, we briefly present the difference between ResNet [11] and DenseNet [12], [32]. The residual path reuses features implicitly, but it does not have any impact on the exploration of new features.…”

Section: B Densely Connected Residual Networkmentioning

confidence: 99%

A Novel Deep Structure U-Net for Sea-Land Segmentation in Remote Sensing Images

Shamsolmoali

Zareapoor

Wang

et al. 2019

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Sea-land segmentation is an important process for many key applications in remote sensing. Proper operative sealand segmentation for remote sensing images remains a challenging issue due to complex and diverse transition between sea and lands. Although several Convolutional Neural Networks (CNNs) have been developed for sea-land segmentation, the performance of these CNNs is far from the expected target. This paper presents a novel deep neural network structure for pixelwise sea-land segmentation, a Residual Dense U-Net (RDU-Net), in complex and high-density remote sensing images. RDU-Net is a combination of both downsampling and upsampling paths to achieve satisfactory results. In each down-and up-sampling path, in addition to the convolution layers, several densely connected residual network blocks are proposed to systematically aggregate multi-scale contextual information. Each dense network block contains multilevel convolution layers, short-range connections and an identity mapping connection which facilitates features reuse in the network and makes full use of the hierarchical features from the original images. These proposed blocks have a certain number of connections that are designed with shorter distance backpropagation between the layers and can significantly improve segmentation results whilst minimizing computational costs. We have performed extensive experiments on two real datasets Google-Earth and ISPRS and compare the proposed RDU-Net against several variations of Dense Networks. The experimental results show that RDU-Net outperforms the other state-of-the-art approaches on the sea-land segmentation tasks.Index Term-Deep neural network (DNN), dense network (DenseNet), remote sensing images, sea-land segmentation, U-Net.

show abstract

“…methods. Therefore, research in the field has long relied on the use of known degradation operators such as bicubic kernel in order to artificially generate a corresponding LR image [9,31,34]. While this straight-forward approach enables simple and efficient benchmarking and generation of virtually unlimited training data, it comes with significant drawbacks.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Learning for Real-World Super-Resolution

Lugmayr

Danelljan

Timofte

2019

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

170

114

View full text Add to dashboard Cite

Most current super-resolution methods rely on low and high resolution image pairs to train a network in a fully supervised manner. However, such image pairs are not available in real-world applications. Instead of directly addressing this problem, most works employ the popular bicubic downsampling strategy to artificially generate a corresponding low resolution image. Unfortunately, this strategy introduces significant artifacts, removing natural sensor noise and other real-world characteristics. Superresolution networks trained on such bicubic images therefore struggle to generalize to natural images.In this work, we propose an unsupervised approach for image super-resolution. Given only unpaired data, we learn to invert the effects of bicubic downsampling in order to restore the natural image characteristics present in the data. This allows us to generate realistic image pairs, faithfully reflecting the distribution of real-world images. Our super-resolution network can therefore be trained with direct pixel-wise supervision in the high resolution domain, while robustly generalizing to real input. We demonstrate the effectiveness of our approach in quantitative and qualitative experiments. arXiv:1909.09629v1 [eess.IV]

show abstract

Image Super-Resolution Using Dense Skip Connections

Cited by 1,092 publications

References 18 publications

Lightweight Image Super-Resolution with Information Multi-distillation Network

Lightweight Image Super-Resolution with Information Multi-distillation Network

A Novel Deep Structure U-Net for Sea-Land Segmentation in Remote Sensing Images

Unsupervised Learning for Real-World Super-Resolution

Contact Info

Product

Resources

About