🌙 DP-Retinex: Dual-Prior Guided Low-Light Image Enhancement with YUV-Domain Reflectance-Illumination Decomposition

Welcome to DP-Retinex! This repository provides a powerful, Retinex-theory-inspired image enhancement model built with PyTorch 🧠. Whether you're working on night-time photography, autonomous vision, or general low-level vision tasks, this toolkit is made for you! 💡

---

🚀 Installation Guide

✅ Step 1: Create a Conda Environment

conda create -n DP-Retinex python=3.8
conda activate DP-Retinex

🧊 Python 3.8 is recommended for compatibility.

---

🔧 Step 2: Install Core Dependencies

pip install torch==1.13.0 torchvision==0.14.0 torchaudio==0.13.0 --extra-index-url https://download.pytorch.org/whl/cu117

🚀 This installs PyTorch with CUDA 11.7 support (for GPU acceleration).

---

📦 Step 3: Install Additional Requirements

pip install -r requirements.txt

This includes packages like numpy, opencv-python, scikit-image, tqdm, etc.

---

🛠️ Step 4: Develop Mode Installation

python setup.py develop --no_cuda_ext

🔧 Use --no_cuda_ext to skip compiling CUDA ops (optional, for environments without NVCC or GPU).

---

📂 Project Structure (简要)

DP-Retinex/
├── basicsr/                      # Core image restoration framework (based on BasicSR)
├── Enhancement/                  # Scripts for evaluation and inference (test_from_dataset.py, etc.)
├── Options/                      # YAML configuration files for different datasets & experiments
├── pretrained_weights/           # (You should manually place downloaded model weights here)

├── data/                         # (You should place LOLv1/LOLv2/SDSD datasets here)
  
├── requirements.txt              # Python dependencies
├── setup.py                      # Package setup file for develop mode
└── README.md                     # You're here 😄

📥 Dataset & Weights Download

🗂️ LOLv1 & LOLv2 Dataset

To evaluate our model, please download the LOLv1 and LOLv2 datasets from the following Google Drive link:

🔗 Download LOLv1 / LOLv2 Dataset (📁 Google Drive)

Please place the datasets under:

data/LOLv1/
data/LOLv2/

---

💾 Pretrained Weights

We provide several pretrained models trained on different benchmarks. You can download all weights from the following Google Drive folder:

🔗 Download Pretrained Weights (📁 Google Drive)

Please organize them as:

pretrained_weights/
│
├── LOLv1_best_psnr_23.48_109000.pth
├── LOLv2_real_best_psnr_23.49_47000.pth
├── LOLv2_syn_bet_psnr_26.30_229000.pth
├── SDSD_indoor_best_psnr_30.70_45000.pth
└── SDSD_outdoor_best_psnr_29.57_60000.pth

---

⚡ Quick Testing

Once your environment is set up and models are downloaded, you can quickly run inference using the following commands:

▶️ LOLv1

python3 Enhancement/test_from_dataset.py \
  --opt Options/Retinex_Degradation_LOL_v1.yml \
  --weights pretrained_weights/LOLv1_best_psnr_23.48_109000.pth

---

▶️ LOLv2

Real subset:

python3 Enhancement/test_from_dataset.py \
  --opt Options/Retinex_Degradation_LOL_v2_real.yml \
  --weights pretrained_weights/LOLv2_real_best_psnr_23.49_47000.pth \
  --dataset LOL_v2_real

Synthetic subset:

python3 Enhancement/test_from_dataset.py \
  --opt Options/Retinex_Degradation_LOL_v2_syn.yml \
  --weights pretrained_weights/LOLv2_syn_bet_psnr_26.30_229000.pth \
  --dataset LOL_v2_syn

---

▶️ SDSD Benchmark

Indoor:

python3 Enhancement/test_from_dataset.py \
  --opt Options/Retinex_Degradation_SDSD_indoorv2.yml \
  --weights pretrained_weights/SDSD_indoor_best_psnr_30.70_45000.pth \
  --dataset SDSD_indoor

Outdoor:

python3 Enhancement/test_from_dataset.py \
  --opt Options/Retinex_Degradation_SDSD_outdoorv2.yml \
  --weights pretrained_weights/SDSD_outdoor_best_psnr_29.57_60000.pth \
  --dataset SDSD_outdoor

---

🧪 Inference Results

All outputs will be saved under:

results/
└── [DatasetName]/
     └── Enhanced/
          ├── image_0001.png
          ├── ...

Each configuration file (*.yml) controls test resolution, padding mode, input/output paths, and model hyperparameters.

---

💬 Tips & Troubleshooting

• ❓ Missing NVCC? Try --no_cuda_ext as above.
• 🧪 Test if PyTorch is installed with GPU:

import torch
print(torch.cuda.is_available())