mlsanityData to 14 Oct 2026

"Zero-Shot" Super-Resolution using Deep Internal Learning

arXiv 1712.06087 ↗First posted 17 Dec 2017cs.CV cs.LG cs.NEcited by 286

Not on the map. Why?

Abstract (arXiv)

Deep Learning has led to a dramatic leap in Super-Resolution (SR) performance in the past few years. However, being supervised, these SR methods are restricted to specific training data, where the acquisition of the low-resolution (LR) images from their high-resolution (HR) counterparts is predetermined (e.g., bicubic downscaling), without any distracting artifacts (e.g., sensor noise, image compression, non-ideal PSF, etc). Real LR images, however, rarely obey these restrictions, resulting in poor SR results by SotA (State of the Art) methods. In this paper we introduce "Zero-Shot" SR, which exploits the power of Deep Learning, but does not rely on prior training. We exploit the internal recurrence of information inside a single image, and train a small image-specific CNN at test time, on examples extracted solely from the input image itself. As such, it can adapt itself to different settings per image. This allows to perform SR of real old photos, noisy images, biological data, and other images where the acquisition process is unknown or non-ideal. On such images, our method outperforms SotA CNN-based SR methods, as well as previous unsupervised SR methods. To the best of our knowledge, this is the first unsupervised CNN-based SR method.

Key facts. Cited by 286 papers in the map (data through 14 Oct 2026). First posted on arXiv 17 Dec 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 74; 2022: 70; 2023: 53; 2024: 46; 2025: 28; 2026: 15.

286 papers.

Sort:

By area:

  1. 2021Designing a Practical Degradation Model for Deep Blind Image Super-Resolutioncited by 529
  2. 2022Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Modelscited by 382
  3. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 359
  4. 2024Learning to (Learn at Test Time): RNNs with Expressive Hidden Statescited by 355
  5. 2022Test-Time Training with Masked Autoencoderscited by 212
  6. 2023Generative Diffusion Prior for Unified Image Restoration and Enhancementcited by 199
  7. 2021Unsupervised Degradation Representation Learning for Blind Super-Resolutioncited by 177
  8. 2021Hand-Object Contact Consistency Reasoning for Human Grasps Generationcited by 160
  9. 2024FeatUp: A Model-Agnostic Framework for Features at Any Resolutioncited by 116
  10. 2021A-SDF: Learning Disentangled Signed Distance Functions for Articulated Shape Representationcited by 114
  11. 2022Real-World Blind Super-Resolution via Feature Matching with Implicit High-Resolution Priorscited by 100
  12. 2023HNeRV: A Hybrid Neural Representation for Videoscited by 100
38 more
  1. 2021Blind Image Super-Resolution: A Survey and Beyondcited by 93
  2. 2022InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Imagescited by 82
  3. 2021Flow-based Kernel Prior with Application to Blind Super-Resolutioncited by 74
  4. 2022Any-resolution Training for High-resolution Image Synthesiscited by 72
  5. 2021Real-World Single Image Super-Resolution: A Brief Reviewcited by 65
  6. 2024Diffusion Models, Image Super-Resolution And Everything: A Surveycited by 64
  7. 2025End-to-End Test-Time Training for Long Contextcited by 62
  8. 2021Zero-Shot Self-Supervised Learning for MRI Reconstructioncited by 54
  9. 2021A Theory of the Distortion-Perception Tradeoff in Wasserstein Spacecited by 52
  10. 2021Mutual Affine Network for Spatially Variant Kernel Estimation in Blind Image Super-Resolutioncited by 52
  11. 2021Drop the GAN: In Defense of Patches Nearest Neighbors as Single Image Generative Modelscited by 51
  12. 2022Deep Constrained Least Squares for Blind Image Super-Resolutioncited by 47
  13. 2022FLEX: Full-Body Grasping Without Full-Body Graspscited by 46
  14. 2023Denoising Diffusion Probabilistic Models for Robust Image Super-Resolution in the Wildcited by 46
  15. 2021Deep Learning for HDR Imaging: State-of-the-Art and Future Trendscited by 45
  16. 2023Image Restoration by Denoising Diffusion Models with Iteratively Preconditioned Guidancecited by 43
  17. 2024Dual Prototype Evolving for Test-Time Generalization of Vision-Language Modelscited by 38
  18. 2021Text Prior Guided Scene Text Image Super-resolutioncited by 37
  19. 2023Test-Time Training on Video Streamscited by 37
  20. 2021Dual-Camera Super-Resolution with Aligned Attention Modulescited by 34
  21. 2021Test-Time Personalization with a Transformer for Human Pose Estimationcited by 33
  22. 2021Blind Image Super-Resolution via Contrastive Representation Learningcited by 32
  23. 2021Removing Diffraction Image Artifacts in Under-Display Camera via Dynamic Skip Connection Networkcited by 31
  24. 2021Deep Matching Prior: Test-Time Optimization for Dense Correspondencecited by 30
  25. 2022Knowledge Distillation based Degradation Estimation for Blind Super-Resolutioncited by 28
  26. 2023CuNeRF: Cube-Based Neural Radiance Field for Zero-Shot Medical Image Arbitrary-Scale Super Resolutioncited by 28
  27. 2021Blind Image Super-resolution with Elaborate Degradation Modeling on Noise and Kernelcited by 27
  28. 2023Recurrence without Recurrence: Stable Video Landmark Detection with Deep Equilibrium Modelscited by 27
  29. 2024LiFT: A Surprisingly Simple Lightweight Feature Transform for Dense ViT Descriptorscited by 27
  30. 2022Ev-TTA: Test-Time Adaptation for Event-Based Object Recognitioncited by 26
  31. 2022Variational Deep Image Restorationcited by 24
  32. 2023Learning Generative Structure Prior for Blind Text Image Super-resolutioncited by 24
  33. 2024You Only Need One Step: Fast Super-Resolution with Stable Diffusion via Scale Distillationcited by 24
  34. 2021A Comprehensive Review of Deep Learning-based Single Image Super-resolutioncited by 23
  35. 2023Neuro-Modulated Hebbian Learning for Fully Test-Time Adaptationcited by 23
  36. 2021Self-Supervised Test-Time Learning for Reading Comprehensioncited by 22
  37. 2022You Only Cut Once: Boosting Data Augmentation with a Single Cutcited by 22
  38. 2022Enriched CNN-Transformer Feature Aggregation Networks for Super-Resolutioncited by 22

50 most cited of 286 All 286 (JSON)