mlsanityData to 14 Oct 2026

Deep Image Prior

arXiv 1711.10925 ↗First posted 29 Nov 2017cs.CV stat.MLcited by 1,098

Not on the map. Why?

Cited by 102 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Deep convolutional networks have become a popular tool for image generation and restoration. Generally, their excellent performance is imputed to their ability to learn realistic image priors from a large number of example images. In this paper, we show that, on the contrary, the structure of a generator network is sufficient to capture a great deal of low-level image statistics prior to any learning. In order to do so, we show that a randomly-initialized neural network can be used as a handcrafted prior with excellent results in standard inverse problems such as denoising, super-resolution, and inpainting. Furthermore, the same prior can be used to invert deep neural representations to diagnose them, and to restore images based on flash-no flash input pairs. Apart from its diverse applications, our approach highlights the inductive bias captured by standard generator network architectures. It also bridges the gap between two very popular families of image restoration methods: learning-based methods using deep convolutional networks and learning-free methods based on handcrafted image priors such as self-similarity. Code and supplementary material are available at https://dmitryulyanov.github.io/deep_image_prior .

Key facts. Cited by 1,098 papers in the map (data through 14 Oct 2026). First posted on arXiv 29 Nov 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 238; 2022: 218; 2023: 196; 2024: 180; 2025: 157; 2026: 109.

1,098 papers.

Sort:

By area:

  1. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,430
  2. 2022Denoising Diffusion Restoration Modelscited by 963
  3. 2021GAN Inversion: A Surveycited by 367
  4. 2023DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Datacited by 337
  5. 2021Text2Mesh: Text-Driven Neural Stylization for Meshescited by 334
  6. 2021Robust Compressed Sensing MRI with Deep Generative Priorscited by 286
  7. 2021NeRV: Neural Representations for Videoscited by 256
  8. 2022Visual Prompting via Image Inpaintingcited by 250
  9. 2022Transformers in Medical Imaging: A Surveycited by 245
  10. 2023WIRE: Wavelet Implicit Neural Representationscited by 220
  11. 2021NeX: Real-time View Synthesis with Neural Basis Expansioncited by 216
  12. 2021A Survey of Human-in-the-loop for Machine Learningcited by 215
38 more
  1. 2022Splicing ViT Features for Semantic Appearance Transfercited by 209
  2. 2023Encoder-based Domain Tuning for Fast Personalization of Text-to-Image Modelscited by 203
  3. 2023Teaching CLIP to Count to Tencited by 198
  4. 2022Neural Feature Fusion Fields: 3D Distillation of Self-Supervised 2D Image Representationscited by 196
  5. 2021Point-Based Neural Rendering with Per-View Optimizationcited by 190
  6. 2021Self-Supervised Learning of Graph Neural Networks: A Unified Reviewcited by 184
  7. 2023Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistencycited by 182
  8. 2021Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Littlecited by 177
  9. 2021Neighbor2Neighbor: Self-Supervised Denoising from Single Noisy Imagescited by 173
  10. 2022SAVi++: Towards End-to-End Object-Centric Learning from Real-World Videoscited by 166
  11. 2022SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Imagecited by 165
  12. 2022RankMe: Assessing the downstream performance of pretrained self-supervised representations by their rankcited by 153
  13. 2021Low-Light Image and Video Enhancement Using Deep Learning: A Surveycited by 148
  14. 2021NeRP: Implicit Neural Representation Learning with Prior Embedding for Sparsely Sampled Image Reconstructioncited by 146
  15. 2021Layered Neural Atlases for Consistent Video Editingcited by 137
  16. 2023RoboCLIP: One Demonstration is Enough to Learn Robot Policiescited by 134
  17. 2023Score-Based Diffusion Models as Principled Priors for Inverse Imagingcited by 127
  18. 2023Generalizing Dataset Distillation via Deep Generative Priorcited by 126
  19. 2022Adaptive Diffusion Priors for Accelerated MRI Reconstructioncited by 118
  20. 2024FeatUp: A Model-Agnostic Framework for Features at Any Resolutioncited by 116
  21. 2021State Entropy Maximization with Random Encoders for Efficient Explorationcited by 108
  22. 2021Gradient Inversion with Generative Image Priorcited by 101
  23. 2023HNeRV: A Hybrid Neural Representation for Videoscited by 100
  24. 2021Blind Image Super-Resolution: A Survey and Beyondcited by 93
  25. 2021Learning to See by Looking at Noisecited by 92
  26. 2021Neural Scene Flow Priorcited by 91
  27. 2023GIR: 3D Gaussian Inverse Rendering for Relightable Scene Factorizationcited by 87
  28. 2022Learning Smooth Neural Functions via Lipschitz Regularizationcited by 86
  29. 2022Blind2Unblind: Self-Supervised Image Denoising with Visible Blind Spotscited by 84
  30. 2023Paint-it: Text-to-Texture Synthesis via Deep Convolutional Texture Map Optimization and Physically-Based Renderingcited by 84
  31. 2021Equivariant Imaging: Learning Beyond the Range Spacecited by 83
  32. 2021Unsupervised MRI Reconstruction via Zero-Shot Learned Adversarial Transformerscited by 77
  33. 2021Flow-based Kernel Prior with Application to Blind Super-Resolutioncited by 74
  34. 2024Provably Robust Score-Based Diffusion Posterior Sampling for Plug-and-Play Image Reconstructioncited by 73
  35. 2024Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizercited by 73
  36. 2023Training-free Linear Image Inverses via Flowscited by 72
  37. 2023Image Denoising: The Deep Learning Revolution and Beyond -- A Survey Paper --cited by 71
  38. 2021High Fidelity Visualization of What Your Self-Supervised Representation Knows Aboutcited by 66

50 most cited of 1,098 All 1,098 (JSON)