mlsanityData to 7 Oct 2026

Adaptive Cross-Layer Attention for Image Restoration

arXiv 2203.03619 ↗First posted 4 Mar 2022eess.IV cs.CV cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Non-local attention module has been proven to be crucial for image restoration. Conventional non-local attention processes features of each layer separately, so it risks missing correlation between features among different layers. To address this problem, we aim to design attention modules that aggregate information from different layers. Instead of finding correlated key pixels within the same layer, each query pixel is encouraged to attend to key pixels at multiple previous layers of the network. In order to efficiently embed such attention design into neural network backbones, we propose a novel Adaptive Cross-Layer Attention (ACLA) module. Two adaptive designs are proposed for ACLA: (1) adaptively selecting the keys for non-local attention at each layer; (2) automatically searching for the insertion locations for ACLA modules. By these two adaptive designs, ACLA dynamically selects a flexible number of keys to be aggregated for non-local attention at previous layer while maintaining a compact neural network with compelling performance. Extensive experiments on image restoration tasks, including single image super-resolution, image denoising, image demosaicing, and image compression artifacts reduction, validate the effectiveness and efficiency of ACLA. The code of ACLA is available at urlhttps://github.com/SDL-ASU/ACLA.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 44 of its 59 references that are in the map. First posted on arXiv 4 Mar 2022.

Builds on

44 of 59 references.

References · 44

  1. 2017Squeeze-and-Excitation Networkscited by 4,763
  2. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,410
  3. 2017Non-local Neural Networkscited by 2,290
  4. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
  5. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,869
  6. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  7. 2021Restormer: Efficient Transformer for High-Resolution Image Restorationcited by 1,618
  8. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  9. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  10. 2017Deformable Convolutional Networkscited by 1,481
  11. 2015Show, Attend and Tell: Neural Image Caption Generation with Visual Attentioncited by 1,421
  12. 2016Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoisingcited by 1,352
32 more
  1. 2018Image Super-Resolution Using Very Deep Residual Channel Attention Networkscited by 1,256
  2. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,218
  3. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  4. 2020Pre-Trained Image Processing Transformercited by 891
  5. 2018Efficient Neural Architecture Search via Parameter Sharingcited by 831
  6. 2018Residual Dense Network for Image Super-Resolutioncited by 796
  7. 2017Progressive Neural Architecture Searchcited by 483
  8. 2017Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolutioncited by 452
  9. 2017Learning Deep CNN Denoiser Prior for Image Restorationcited by 407
  10. 2017Residual Attention Network for Image Classificationcited by 385
  11. 2019Attention Augmented Convolutional Networkscited by 363
  12. 2015Sketch-based Manga Retrieval using Manga109 Datasetcited by 326
  13. 2018SNAS: Stochastic Neural Architecture Searchcited by 288
  14. 2020Single Image Super-Resolution via a Holistic Attention Networkcited by 265
  15. 2018Fast, Accurate, and Lightweight Super-Resolution with Cascading Residual Networkcited by 263
  16. 2019Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentationcited by 258
  17. 2017MemNet: A Persistent Memory Network for Image Restorationcited by 244
  18. 2019Residual Non-local Attention Networks for Image Restorationcited by 236
  19. 2017Learning a Single Convolutional Super-Resolution Network for Multiple Degradationscited by 227
  20. 2018Deep Back-Projection Networks For Super-Resolutioncited by 224
  21. 2016SCA-CNN: Spatial and Channel-wise Attention in Convolutional Networks for Image Captioningcited by 219
  22. 2018Non-Local Recurrent Network for Image Restorationcited by 160
  23. 2015Compression Artifacts Reduction by a Deep Convolutional Networkcited by 136
  24. 2017Genetic CNNcited by 124
  25. 2019Dynamic Convolutions: Exploiting Spatial Sparsity for Faster Inferencecited by 91
  26. 2016A survey of sparse representation: algorithms and applicationscited by 58
  27. 2019Densely Connected Search Space for More Flexible Neural Architecture Searchcited by 52
  28. 2020DCNAS: Densely Connected Neural Architecture Search for Semantic Image Segmentationcited by 33
  29. 2021OmniNet: Omnidirectional Representations from Transformerscited by 23
  30. 2020Hierarchical Neural Architecture Search for Single Image Super-Resolutioncited by 18
  31. 2020Pyramid Attention Networks for Image Restorationcited by 17
  32. 2019Scale-wise Convolution for Image Restorationcited by 14

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 1; 2025: 1.

3 papers.

  1. 2023Cross-Layer Retrospective Retrieving via Layer Attentioncited by 20
  2. 2024Efficient Visual Transformer by Learnable Token Mergingcited by 7
  3. 2025Compact Vision Transformer by Reduction of Kernel Complexitycited by 0