• Corpus ID: 220265400

Cross-Scale Internal Graph Neural Network for Image Super-Resolution

  title={Cross-Scale Internal Graph Neural Network for Image Super-Resolution},
  author={Shangchen Zhou and Jiawei Zhang and Wangmeng Zuo and Chen Change Loy},
Non-local self-similarity in natural images has been well studied as an effective prior in image restoration. However, for single image super-resolution (SISR), most existing deep non-local methods (e.g., non-local neural networks) only exploit similar patches within the same scale of the low-resolution (LR) input image. Consequently, the restoration is limited to using the same-scale information while neglecting potential high-resolution (HR) cues from other scales. In this paper, we explore… 

Figures and Tables from this paper

Learning Frequency-aware Dynamic Network for Efficient Super-Resolution
This paper explores a novel frequency-aware dynamic network for dividing the input into multiple parts according to its coefficients in the discrete cosine transform (DCT) domain and can be employed for various SISR neural architectures to obtain the better tradeoff between visual quality and computational complexity.
Hyperrealistic Image Inpainting with Hypergraphs
This paper introduces a trainable mechanism to connect nodes using hyperedges for hypergraph convolution and introduces gated convolution in the discriminator to enforce local consistency in the predicted image.
GLEAN: Generative Latent Bank for Large-Factor Image Super-Resolution
This work shows that pre-trained Generative Adversarial Networks (GANs), e.g., StyleGAN, can be used as a latent bank to improve the restoration quality of large-factor image super-resolution (SR) and shows clear improvements in terms of fidelity and texture faithfulness in comparison to existing methods.
Pre-Trained Image Processing Transformer
To maximally excavate the capability of transformer, the IPT model is presented to utilize the well-known ImageNet benchmark for generating a large amount of corrupted image pairs and the contrastive learning is introduced for well adapting to different image processing tasks.
Holistic 3D Scene Understanding from a Single Image with Implicit Representation
This work proposes an image-based local structured implicit network to improve the object shape estimation, but also refine the 3D object pose and scene layout via a novel implicit scene graph neural network that exploits the implicit local object features.
Rich CNN-Transformer Feature Aggregation Networks for Super-Resolution
An effective hybrid architecture for super-resolution (SR) tasks, which leverages local features from CNNs and long-range dependencies captured by transformers to further improve the SR results and proposes a cross-scale token attention module, which allows the transformer to efficiently exploit the informative relationships among tokens across different scales.
SwinIR: Image Restoration Using Swin Transformer
A strong baseline model SwinIR is proposed for image restoration based on the Swin Transformer that outperforms state-of-the-art methods on different tasks by up to 0.14∼0.45dB, while the total number of parameters can be reduced byUp to 67%.
Learning Graph Neural Networks for Image Style Transfer
Experimental results demonstrate that the proposed semi-parametric image stylization approach yields encouraging results on the challenging style patterns, preserving both global appearance and exquisite details.
Boosting Video Super Resolution with Patch-Based Temporal Redundancy Optimization
Two simple plug and play methods to improve the performance of existing local and non-local propagation-based VSR algorithms on widely-used public videos by handling the temporal redundancy patches in an optimized manner are developed.
Towards Lightweight Super-Resolution with Dual Regression Learning
A lightweight dual regression compression method to reduce model redundancy in both layer-level and channel-level based on channel pruning is proposed to address the first challenge and the effectiveness of the method is shown in obtaining accurate and efficient SR models.


Fast Image Super-Resolution Based on In-Place Example Regression
We propose a fast regression model for practical single image super-resolution based on in-place examples, by leveraging two fundamental super-resolution approaches- learning from an external
Deep Graph-Convolutional Image Denoising
This paper proposes a novel end-to-end trainable neural network architecture employing layers based on graph convolution operations, thereby creating neurons with non-local receptive fields and introduces a lightweight Edge-Conditioned Convolution which addresses vanishing gradient and over-parameterization issues of this particulargraph convolution.
Non-Local Recurrent Network for Image Restoration
A non- local recurrent network (NLRN) is proposed as the first attempt to incorporate non-local operations into a recurrent neural network (RNN) for image restoration and achieves superior results to state-of-the-art methods with much fewer parameters.
Single image super-resolution from transformed self-exemplars
This paper expands the internal patch search space by allowing geometric variations, and proposes a compositional model to simultaneously handle both types of transformations to accommodate local shape variations.
Super-resolution through neighbor embedding
  • Hong Chang, D. Yeung, Yimin Xiong
  • Computer Science, Mathematics
    Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.
  • 2004
This paper proposes a novel method for solving single-image super-resolution problems, given a low-resolution image as input, and recovers its high-resolution counterpart using a set of training examples, inspired by recent manifold teaming methods.
Neural Nearest Neighbors Networks
The neural nearest neighbors block (N3 block), a novel non-local processing layer that leverages the principle of self-similarity and can be used as building block in modern neural network architectures is proposed, showing its effectiveness for the set reasoning task of correspondence classification and for image restoration.
Low-Complexity Single-Image Super-Resolution based on Nonnegative Neighbor Embedding
The neighbor embedding SR algorithm so designed is shown to give good visual results, comparable to other state-of-the-art methods, while presenting an appreciable reduction of the computational time.
Second-Order Attention Network for Single Image Super-Resolution
Experimental results demonstrate the superiority of the SAN network over state-of-the-art SISR methods in terms of both quantitative metrics and visual quality.
Deep Networks for Image Super-Resolution with Sparse Prior
This paper shows that a sparse coding model particularly designed for super-resolution can be incarnated as a neural network, and trained in a cascaded structure from end to end, and leads to much more efficient and effective training, as well as a reduced model size.
Residual Dense Network for Image Super-Resolution
This paper proposes residual dense block (RDB) to extract abundant local features via dense connected convolutional layers and uses global feature fusion in RDB to jointly and adaptively learn global hierarchical features in a holistic way.