# dmitryulyanov/deep-image-prior

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/dmitryulyanov-deep-image-prior).**

8,085 stars · 1,439 forks · Jupyter Notebook · NOASSERTION

## Links

- GitHub: https://github.com/DmitryUlyanov/deep-image-prior
- Homepage: https://dmitryulyanov.github.io/deep_image_prior
- awesome-repositories: https://awesome-repositories.com/repository/dmitryulyanov-deep-image-prior.md

## Description

This project is an unsupervised image restoration tool that uses a convolutional neural network as a structural prior to reconstruct images from noisy or incomplete data. It functions as a neural network image prior, utilizing the inherent biases of the network architecture to restore pixels without the need for a pre-trained dataset or external learning.

The system performs zero-shot image restoration by treating the network architecture itself as a regularization term. It uses a randomly initialized encoder-decoder structure and iterative gradient descent to minimize pixel-wise loss, recovering the natural image structure through self-supervised reconstruction.

The tool covers several image recovery domains, including image denoising, super-resolution, and general image restoration. It also provides capabilities for blind image deconvolution to recover sharp images from blurred versions.

## Tags

### Artificial Intelligence & ML

- [Image Restorers](https://awesome-repositories.com/f/artificial-intelligence-ml/convolutional-neural-networks/image-restorers.md) — Provides a convolutional neural network based tool for unsupervised image denoising and super-resolution.
- [Structural Priors](https://awesome-repositories.com/f/artificial-intelligence-ml/convolutional-neural-networks/structural-priors.md) — Implements a convolutional neural network as a structural prior to restore images without needing external training datasets.
- [Self-Supervised Reconstruction](https://awesome-repositories.com/f/artificial-intelligence-ml/image-region-reconstruction/self-supervised-reconstruction.md) — Performs image restoration by treating the neural network architecture as a self-supervised regularization term.
- [Image Restoration Models](https://awesome-repositories.com/f/artificial-intelligence-ml/image-restoration-models.md) — Repairs corrupted image data and fills missing pixels by optimizing a reconstruction network.
- [Unsupervised Restoration Models](https://awesome-repositories.com/f/artificial-intelligence-ml/image-restoration-models/unsupervised-restoration-models.md) — Recovers corrupted pixels using an unsupervised approach that requires no pre-trained dataset.
- [Image Super Resolution Models](https://awesome-repositories.com/f/artificial-intelligence-ml/image-super-resolution-models.md) — Increases image resolution and restores fine details by leveraging the structural properties of a convolutional network.
- [Iterative Image Optimizers](https://awesome-repositories.com/f/artificial-intelligence-ml/iterative-image-optimizers.md) — Refines reconstructed image pixels through iterative gradient descent to minimize pixel-wise loss.
- [Generative Priors](https://awesome-repositories.com/f/artificial-intelligence-ml/probabilistic-priors/generative-priors.md) — Leverages the inherent bias of convolutional layers as a structural prior for image reconstruction.
- [L1 Pixel Loss](https://awesome-repositories.com/f/artificial-intelligence-ml/adversarial-loss-functions/l1-pixel-loss.md) — Minimizes the L1 pixel-wise difference between the synthesized output and the corrupted target image.
- [Encoder-Decoder Architectures](https://awesome-repositories.com/f/artificial-intelligence-ml/encoder-decoder-architectures.md) — Utilizes an encoder-decoder structure to capture multiscale spatial information for image reconstruction.
- [Weight Initialization](https://awesome-repositories.com/f/artificial-intelligence-ml/weight-initialization.md) — Uses random weight initialization to ensure the network recovers natural image structures through the optimization process.

### Part of an Awesome List

- [Zero-Shot Restoration](https://awesome-repositories.com/f/awesome-lists/ai/image-restoration/zero-shot-restoration.md) — Performs zero-shot image restoration by applying a neural network architecture as a prior without pre-trained datasets. ([source](https://github.com/dmitryulyanov/deep-image-prior#readme))
- [Computer Vision Models](https://awesome-repositories.com/f/awesome-lists/ai/computer-vision-models.md) — Restores images using neural networks without requiring prior learning.
- [Computer Vision Research](https://awesome-repositories.com/f/awesome-lists/ai/computer-vision-research.md) — Unsupervised image restoration using neural network architecture priors.
- [Image Inpainting Models](https://awesome-repositories.com/f/awesome-lists/ai/image-inpainting-models.md) — Unsupervised image restoration using deep generative priors.
- [Non-Blind Deblurring](https://awesome-repositories.com/f/awesome-lists/ai/non-blind-deblurring.md) — Uses deep image priors for various restoration tasks without pre-training.

### Graphics & Multimedia

- [Image Denoising](https://awesome-repositories.com/f/graphics-multimedia/image-denoising.md) — Removes unwanted noise from digital images using a neural network prior without training data.
- [Blind Image Deconvolution](https://awesome-repositories.com/f/graphics-multimedia/blind-image-deconvolution.md) — Provides capabilities for blind image deconvolution to recover sharp images from blurred versions.
- [Joint Denoising and Super-Resolution](https://awesome-repositories.com/f/graphics-multimedia/joint-denoising-and-super-resolution.md) — Combines image denoising and super-resolution capabilities using a randomly initialized network prior. ([source](https://dmitryulyanov.github.io/deep_image_prior))