mlsanityData to 14 Oct 2026

Reflash Dropout in Image Super-Resolution

arXiv 2112.12089 ↗First posted 22 Dec 2021cs.CVcited by 41

Not on the map. Why?

Abstract (arXiv)

Dropout is designed to relieve the overfitting problem in high-level vision tasks but is rarely applied in low-level vision tasks, like image super-resolution (SR). As a classic regression problem, SR exhibits a different behaviour as high-level tasks and is sensitive to the dropout operation. However, in this paper, we show that appropriate usage of dropout benefits SR networks and improves the generalization ability. Specifically, dropout is better embedded at the end of the network and is significantly helpful for the multi-degradation settings. This discovery breaks our common sense and inspires us to explore its working mechanism. We further use two analysis tools -- one is from recent network interpretation works, and the other is specially designed for this task. The analysis results provide side proofs to our experimental findings and show us a new perspective to understand SR networks.

Key facts. Cited by 41 papers in the map (data through 14 Oct 2026). Builds on 41 of its 65 references that are in the map. First posted on arXiv 22 Dec 2021.

Builds on

41 of 65 references.

References · 41

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  3. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  4. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  5. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  6. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,908
  7. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,887
  8. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,552
  9. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  10. 2018Image Super-Resolution Using Very Deep Residual Channel Attention Networkscited by 1,264
  11. 2018ESRGAN: Enhanced Super-Resolution Generative Adversarial Networkscited by 1,254
  12. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,224
29 more
  1. 2014Striving for Simplicity: The All Convolutional Netcited by 990
  2. 2021Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Datacited by 970
  3. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 903
  4. 2018Residual Dense Network for Image Super-Resolutioncited by 799
  5. 2021Designing a Practical Degradation Model for Deep Blind Image Super-Resolutioncited by 529
  6. 2015Sketch-based Manga Retrieval using Manga109 Datasetcited by 329
  7. 2018DropBlock: A regularization method for convolutional networkscited by 262
  8. 2014Efficient Object Localization Using Convolutional Networkscited by 227
  9. 2016FractalNet: Ultra-Deep Neural Networks without Residualscited by 216
  10. 2019Blind Super-Resolution With Iterative Kernel Correctioncited by 203
  11. 2013Maxout Networkscited by 187
  12. 2019Image Processing Using Multi-Code GAN Priorcited by 185
  13. 2020PIPAL: a Large-Scale Image Quality Assessment Dataset for Perceptual Image Restorationcited by 178
  14. 2021Unsupervised Degradation Representation Learning for Blind Super-Resolutioncited by 177
  15. 2019Blind Super-Resolution Kernel Estimation using an Internal-GANcited by 175
  16. 2019RankSRGAN: Generative Adversarial Networks with Ranker for Image Super-Resolutioncited by 145
  17. 2020Interpreting Super-Resolution Networks with Local Attribution Mapscited by 136
  18. 2020Unfolding the Alternating Optimization for Blind Super Resolutioncited by 129
  19. 2018On the importance of single directions for generalizationcited by 99
  20. 2021Blind Image Super-Resolution: A Survey and Beyondcited by 93
  21. 2020Efficient Image Super-Resolution Using Pixel Attentioncited by 87
  22. 2021ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristiccited by 81
  23. 2020Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Analysis and a New Strategycited by 66
  24. 2015Dropout as data augmentationcited by 27
  25. 2021Finding Discriminative Filters for Specific Degradations in Blind Super-Resolutioncited by 26
  26. 2021Attention in Attention Network for Image Super-Resolutioncited by 20
  27. 2014Dropout Rademacher Complexity of Deep Neural Networkscited by 14
  28. 2019Suppressing Model Overfitting for Image Super-Resolution Networkscited by 9
  29. 2015To Drop or Not to Drop: Robustness, Consistency and Differential Privacy Properties of Dropoutcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 13; 2023: 10; 2024: 10; 2025: 6; 2026: 2.

41 papers.

Sort:

By area:

  1. 2022Activating More Pixels in Image Super-Resolution Transformercited by 438
  2. 2024Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wildcited by 286
  3. 2022Cross Aggregation Transformer for Image Restorationcited by 92
  4. 2022Accurate Image Restoration with Attention Retractable Transformercited by 85
  5. 2022NAFSSR: Stereo Image Super-Resolution Using NAFNetcited by 76
  6. 2023HAT: Hybrid Attention Transformer for Image Restorationcited by 66
  7. 2024I2EBench: A Comprehensive Benchmark for Instruction-based Image Editingcited by 63
  8. 2022MogaNet: Multi-order Gated Aggregation Networkcited by 59
  9. 2023Masked Image Training for Generalizable Deep Image Denoisingcited by 57
  10. 2023MTS-Mixers: Multivariate Time Series Forecasting via Factorized Temporal and Channel Mixingcited by 53
  11. 2022N-Gram in Swin Transformers for Efficient Lightweight Image Super-Resolutioncited by 46
  12. 2022Towards Practical Certifiable Patch Defense with Vision Transformercited by 44
29 more
  1. 2024An Intelligent Agentic System for Complex Image Restoration Problemscited by 43
  2. 2023Crafting Training Degradation Distribution for the Accuracy-Generalization Trade-off in Real-World Super-Resolutioncited by 21
  3. 2025Fine-structure Preserved Real-world Image Super-resolution via Transfer VAE Trainingcited by 21
  4. 2022NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video: Dataset, Methods and Resultscited by 20
  5. 2022Evaluating the Generalization Ability of Super-Resolution Networkscited by 18
  6. 2024Beyond Pixels: Text Enhances Generalization in Real-World Image Restorationcited by 17
  7. 2023Networks are Slacking Off: Understanding Generalization Problem in Image Derainingcited by 14
  8. 2022Towards Lightweight Super-Resolution with Dual Regression Learningcited by 13
  9. 2023SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolutioncited by 12
  10. 2022UDC-UNet: Under-Display Camera Image Restoration via U-Shape Dynamic Networkcited by 11
  11. 2023SwinFSR: Stereo Image Super-Resolution using SwinIR and Frequency Domain Knowledgecited by 10
  12. 2024Blind Super-Resolution via Meta-learning and Markov Chain Monte Carlo Simulationcited by 7
  13. 2025MRI Super-Resolution with Deep Learning: A Comprehensive Surveycited by 7
  14. 2024A Preliminary Exploration Towards General Image Restorationcited by 6
  15. 2025Revisiting the Generalization Problem of Low-level Vision Models Through the Lens of Image Derainingcited by 4
  16. 2022Super-Resolution by Predicting Offsets: An Ultra-Efficient Super-Resolution Network for Rasterized Imagescited by 3
  17. 2023SuperInpaint: Learning Detail-Enhanced Attentional Implicit Representation for Super-resolutional Image Inpaintingcited by 3
  18. 2024Navigating Beyond Dropout: An Intriguing Solution Towards Generalizable Image Super Resolutioncited by 3
  19. 2025UniCon: Unidirectional Information Flow for Effective Control of Large-Scale Diffusion Modelscited by 3
  20. 2022ITSRN++: Stronger and Better Implicit Transformer Network for Continuous Screen Content Image Super-Resolutioncited by 2
  21. 2023StarSRGAN: Improving Real-World Blind Super-Resolutioncited by 2
  22. 2024Federated Learning for Blind Image Super-Resolutioncited by 2
  23. 2024BatSort: Enhanced Battery Classification with Transfer Learning for Battery Sorting and Recyclingcited by 1
  24. 2026Restoration Adaptation for Semantic Segmentation on Low Quality Imagescited by 1
  25. 2023I2V: Towards Texture-Aware Self-Supervised Blind Denoising using Self-Residual Learning for Real-World Imagescited by 0
  26. 2024Sewer Image Super-Resolution with Depth Priors and Its Lightweight Networkcited by 0
  27. 2025Beyond Overfitting: Doubly Adaptive Dropout for Generalizable AU Detectioncited by 0
  28. 2025Not All Degradations Are Equal: A Targeted Feature Denoising Framework for Generalizable Image Super-Resolutioncited by 0
  29. 2026An Extensive Benchmark for Single-round and Multi-round Instruction-based Image Editingcited by 0