Conditional-Consistency-Models

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #typescript #golang #android #rust #linux #java #nodejs #windows #vue #awesome-list

Conditional-Consistency-Models

(★ 18)

This repository contains the official implementation and supplementary materials for the research paper "Conditional Consistency Guided Image Translation and Enhancement". It explores advanced consistency models for conditional data generation, focusing on tasks like low-light enhancement and infrared-to-visible translation.

Conditional-Consistency-Models Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Conditional Consistency Models

Welcome to the official repository for Conditional Consistency Models (CCM). This repository hosts the implementation and evaluation of consistency models tailored for various datasets, including **IRVI**, **BCI**, **LLVIP**, **LOLv1**, **LOLv2**, and **SID**.

This repository contains code for training and evaluating models.

## Table of Contents

1. [Requirements and Setup](#requirements-and-setup)
2. [Directory Structure](#directory-structure)
3. [Datasets](#datasets)
4. [Usage](#usage)

## Requirements and Setup

### 1. Clone the Repository
```bash
cd Conditional-Consistency-Models
```

### 2. Create a Conda Environment
Create and activate a new conda environment named ccm:
```bash
conda create -n ccm python=3.10 -y
conda activate ccm
```

### 3. Install Dependencies
Install all required Python packages:
```bash
pip install -r requirements.txt
```

## Directory Structure

Ensure your project directory is organized as follows:

```
Conditional-Consistency-Models/
    ├── datasets/                 # Folder for datasets
    │   ├── bci/
    │   ├── llvip/
    │   ├── lolv1/
    │   ├── lolv2/
    │   ├── sid/
    │   ├── irvi/
    ├── bci/                      # Folder for BCI model scripts and metrics
    │   ├── script.py
    │   ├── sampling_and_metric.py
    │   ├── checkpoints/bci/          # Pre-trained BCI model weights
    │       ├── model.json
    │       ├── model.pt
    ├── llvip/                   
    ├── lolv1/
    ├── lolv2/
    ├── sid/
    ├── irvi/
    ├── improved_consistency_model_conditional.py
    ├── README.md
    ├── requirements.txt
```

## Datasets

### Download Datasets

Use the links below to download datasets for each model. Once downloaded, extract and place them inside the `datasets/` directory.

- BCI Dataset: [https://bupt-ai-cz.github.io/BCI/](URL)
- LLVIP Dataset: [https://drive.google.com/file/d/1VTlT3Y7e1h-Zsne4zahjx5q0TK2ClMVv/view](URL)
- LOLv1 Dataset: [https://drive.google.com/file/d/1L-kqSQyrmMueBh_ziWoPFhfsAh50h20H/view](URL)
- LOLv2 Dataset: [https://drive.google.com/file/d/1Ou9EljYZW8o5dbDCf9R34FS8Pd8kEp2U/view](URL)
- SID Dataset: [https://drive.google.com/drive/folders/1eQ-5Z303sbASEvsgCBSDbhijzLTWQJtR](URL)
- IRVI Dataset: [https://drive.google.com/file/d/1ZcJ0EfF5n_uqtsLc7-8hJgTcr2zHSXY3/view](URL)

## Usage

### 1. Training

To train a model on a specific dataset, use the corresponding `script.py` inside the dataset's folder.

For example, to train the BCI model:
```bash
cd ..
CUDA_VISIBLE_DEVICES=0 python -m bci.script
```

### 2. Evaluation (Metrics)

To evaluate the model and calculate PSNR/SSIM metrics, use the `sampling_and_metric.py` script.

Example: Evaluate the BCI model:
```bash
cd ..
CUDA_VISIBLE_DEVICES=0 python -m bci.sampling_and_metric
```

Ensure the dataset is correctly placed, and pre-trained checkpoints are available before running the evaluation.

### Results

Evaluation metrics (PSNR, SSIM) and generated images are saved in the corresponding output folder.

For any issues, feel free to open an issue on this repository.

Conditional-Consistency-Models

Conditional-Consistency-Models Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Conditional-Consistency-Models

Conditional-Consistency-Models Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README