• Corpus ID: 244714326

AdaDM: Enabling Normalization for Image Super-Resolution

  title={AdaDM: Enabling Normalization for Image Super-Resolution},
  author={Jie Liu and Jie Tang and Gangshan Wu},
Normalization like Batch Normalization (BN) is a milestone technique to normalize the distributions of intermediate layers in deep learning, enabling faster training and better generalization accuracy. However, in fidelity image Super-Resolution (SR), it is believed that normalization layers get rid of range flexibility by normalizing the features and they are simply removed from modern SR networks. In this paper, we study this phenomenon quantitatively and qualitatively. We found that the… 


Group Normalization
Group Normalization can outperform its BN-based counterparts for object detection and segmentation in COCO, and for video classification in Kinetics, showing that GN can effectively replace the powerful BN in a variety of tasks.
Accelerating the Super-Resolution Convolutional Neural Network
This paper aims at accelerating the current SRCNN, and proposes a compact hourglass-shape CNN structure for faster and better SR, and presents the parameter settings that can achieve real-time performance on a generic CPU while still maintaining good performance.
Enhanced Deep Residual Networks for Single Image Super-Resolution
This paper develops an enhanced deep super-resolution network (EDSR) with performance exceeding those of current state-of-the-art SR methods, and proposes a new multi-scale deepsuper-resolution system (MDSR) and training method, which can reconstruct high-resolution images of different upscaling factors in a single model.
Feedback Network for Image Super-Resolution
An image super-resolution feedback network (SRFBN) is proposed to refine low-level representations with high-level information by using hidden states in a recurrent neural network (RNN) with constraints to achieve such feedback manner.
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
This work thoroughly study three key components of SRGAN – network architecture, adversarial loss and perceptual loss, and improves each of them to derive an Enhanced SRGAN (ESRGAN), which achieves consistently better visual quality with more realistic and natural textures than SRGAN.
Learning with Privileged Information for Efficient Image Super-Resolution
This paper introduces a novel distillation framework, consisting of teacher and student networks, that allows to boost the performance of FSRCNN drastically, and proposes to use ground-truth high-resolution (HR) images as privileged information.
Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution
This paper proposes the Laplacian Pyramid Super-Resolution Network (LapSRN) to progressively reconstruct the sub-band residuals of high-resolution images and generates multi-scale predictions in one feed-forward pass through the progressive reconstruction, thereby facilitates resource-aware applications.
Learning a Single Convolutional Super-Resolution Network for Multiple Degradations
  • K. Zhang, W. Zuo, Lei Zhang
  • Computer Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • 2018
Extensive experimental results show that the proposed convolutional super-resolution network not only can produce favorable results on multiple degradations but also is computationally efficient, providing a highly effective and scalable solution to practical SISR applications.
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.
Residual Feature Aggregation Network for Image Super-Resolution
This work proposes a novel residual feature aggregation (RFA) framework for more efficient feature extraction and proposes an enhanced spatial attention (ESA) block to make the residual features to be more focused on critical spatial contents.