# enrico-favale/deep-image-restoration architecture

> Architecture diagram of the GitHub repository enrico-favale/deep-image-restoration, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-21). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/enrico-favale/deep-image-restoration
- Repository: https://github.com/enrico-favale/deep-image-restoration

## Overview

This is a research codebase for restoring Fourier- and wavelet-degraded BSD500 images with convolutional autoencoders and Restormer. Its principal workflow is to create compressed images, pair them with originals, train a restoration model, run inference, and compare restoration quality with image metrics. The graph also shows training metadata and visualization as supporting capabilities. Some relationships are documented only at a high level in the README; unsampled notebook wiring and implementation details are not asserted.

Main entry points: [`src/utils/trainer.py`](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/trainer.py) (Model trainer), [`src/compression/fourier.py`](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/fourier.py) (Fourier compression), [`src/compression/wavelet.py`](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/wavelet.py) (Wavelet compression), [`src/utils/inference.py`](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/inference.py) (Image restoration)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_compression["Image compression"]
  node_fourier["Fourier compression<br/>[fourier.py]"]
  node_wavelet["Wavelet compression<br/>[wavelet.py]"]
  node_compressor_base["Compressor interface<br/>[compressor.py]"]
  node_compression_utils["Image file utilities<br/>[utils.py]"]
end

subgraph group_learning["Restoration learning"]
  node_pair_dataset["Image-pair dataset<br/>[preprocessing.py]"]
  node_trainer["Model trainer<br/>[trainer.py]"]
  node_autoencoder["Convolutional autoencoder<br/>[autoencoder.py]"]
  node_restormer["Restormer<br/>[restormer.py]"]
end

subgraph group_assessment["Inference and assessment"]
  node_inference["Image restoration<br/>[inference.py]"]
  node_evaluation["Quality evaluation<br/>[evaluation.py]"]
  node_visualization["Restoration plots<br/>[plot.py]"]
end

subgraph group_support["Training support"]
  node_logger["Experiment logger<br/>[logger.py]"]
end

node_researcher(("Researcher"))
node_source_images["Original images"]
node_compressed_images["Compressed images"]
node_checkpoints["Model checkpoints"]
node_restored_image["Restored image"]

node_researcher -.->|"compresses with"| node_fourier
node_researcher -.->|"compresses with"| node_wavelet
node_source_images -->|"provides images"| node_fourier
node_source_images -->|"provides images"| node_wavelet
node_fourier -->|"implements"| node_compressor_base
node_wavelet -->|"implements"| node_compressor_base
node_fourier -->|"loads and saves"| node_compression_utils
node_wavelet -->|"loads and saves"| node_compression_utils
node_fourier -->|"produces"| node_compressed_images
node_wavelet -->|"produces"| node_compressed_images
node_compressed_images -->|"pairs with originals"| node_pair_dataset
node_source_images -->|"provides targets"| node_pair_dataset
node_researcher -.->|"trains model"| node_trainer
node_trainer -->|"loads and splits"| node_pair_dataset
node_trainer -->|"trains supplied model"| node_autoencoder
node_trainer -->|"records experiment"| node_logger
node_autoencoder -->|"logs architecture"| node_logger
node_trainer -.->|"saves best model"| node_checkpoints
node_researcher -.->|"restores image"| node_inference
node_checkpoints -->|"provides weights"| node_inference
node_inference -->|"runs supplied model"| node_autoencoder
node_inference -.->|"runs supplied model"| node_restormer
node_inference -->|"returns image"| node_restored_image
node_researcher -.->|"evaluates quality"| node_evaluation
node_evaluation -.->|"measures restoration"| node_restored_image
node_researcher -.->|"compares results"| node_visualization

click node_fourier "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/fourier.py"
click node_wavelet "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/wavelet.py"
click node_compressor_base "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/compressor.py"
click node_compression_utils "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/utils.py"
click node_pair_dataset "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/preprocessing.py"
click node_trainer "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/trainer.py"
click node_autoencoder "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/models/autoencoder.py"
click node_restormer "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/models/restormer.py"
click node_inference "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/inference.py"
click node_evaluation "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/evaluation.py"
click node_visualization "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/plot.py"
click node_logger "https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/logger.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_fourier,node_wavelet,node_compressor_base,node_compression_utils toneBlue
class node_pair_dataset,node_trainer,node_autoencoder,node_restormer toneAmber
class node_inference,node_evaluation,node_visualization toneMint
class node_logger toneRose
class node_researcher,node_source_images,node_compressed_images,node_checkpoints,node_restored_image toneIndigo
```

## Components

### Image compression

- [Fourier compression](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/fourier.py): `src/compression/fourier.py`
- [Wavelet compression](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/wavelet.py): `src/compression/wavelet.py`
- [Compressor interface](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/compressor.py): `src/compression/compressor.py`
- [Image file utilities](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/compression/utils.py): `src/compression/utils.py`

### Restoration learning

- [Image-pair dataset](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/preprocessing.py): `src/utils/preprocessing.py`
- [Model trainer](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/trainer.py): `src/utils/trainer.py`
- [Convolutional autoencoder](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/models/autoencoder.py): `src/models/autoencoder.py`
- [Restormer](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/models/restormer.py): `src/models/restormer.py`

### Inference and assessment

- [Image restoration](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/inference.py): `src/utils/inference.py`
- [Quality evaluation](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/evaluation.py): `src/utils/evaluation.py`
- [Restoration plots](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/plot.py): `src/utils/plot.py`

### Training support

- [Experiment logger](https://github.com/enrico-favale/deep-image-restoration/blob/main/src/utils/logger.py): `src/utils/logger.py`

### Other components

- **Researcher**
- **Original images**
- **Compressed images**
- **Model checkpoints**
- **Restored image**

## Connections

- Researcher → Fourier compression: compresses with (evidence: `README.md`)
- Researcher → Wavelet compression: compresses with (evidence: `README.md`)
- Original images → Fourier compression: provides images (evidence: `src/compression/fourier.py`)
- Original images → Wavelet compression: provides images (evidence: `src/compression/wavelet.py`)
- Fourier compression → Compressor interface: implements (evidence: `src/compression/fourier.py`)
- Wavelet compression → Compressor interface: implements (evidence: `src/compression/wavelet.py`)
- Fourier compression → Image file utilities: loads and saves (evidence: `src/compression/fourier.py`)
- Wavelet compression → Image file utilities: loads and saves (evidence: `src/compression/wavelet.py`)
- Fourier compression → Compressed images: produces (evidence: `src/compression/fourier.py`)
- Wavelet compression → Compressed images: produces (evidence: `src/compression/wavelet.py`)
- Compressed images → Image-pair dataset: pairs with originals (evidence: `src/utils/preprocessing.py`)
- Original images → Image-pair dataset: provides targets (evidence: `src/utils/preprocessing.py`)
- Researcher → Model trainer: trains model (evidence: `README.md`)
- Model trainer → Image-pair dataset: loads and splits (evidence: `src/utils/trainer.py`)
- Model trainer → Convolutional autoencoder: trains supplied model (evidence: `src/utils/trainer.py`)
- Model trainer → Experiment logger: records experiment (evidence: `src/utils/trainer.py`)
- Convolutional autoencoder → Experiment logger: logs architecture (evidence: `src/models/autoencoder.py`)
- Model trainer → Model checkpoints: saves best model
- Researcher → Image restoration: restores image (evidence: `README.md`)
- Model checkpoints → Image restoration: provides weights (evidence: `src/utils/inference.py`)
- Image restoration → Convolutional autoencoder: runs supplied model (evidence: `src/utils/inference.py`)
- Image restoration → Restormer: runs supplied model
- Image restoration → Restored image: returns image (evidence: `src/utils/inference.py`)
- Researcher → Quality evaluation: evaluates quality (evidence: `README.md`)
- Quality evaluation → Restored image: measures restoration
- Researcher → Restoration plots: compares results
