LOL Dataset

Estimated reading: 3 minutes 2792 views

Visualization of the LOL Train Dataset in the Deep Lake UI

LOL dataset

What is LOL Dataset?

The LOL (Low-Light) dataset is a benchmark dataset designed to address the real-world challenge of low-light image enhancement. The LOL dataset is divided into 485 training pairs and 15 testing pairs and contains 500 low-light and normal-light image pairs. The noise in the low-light image was created during the photo-taking procedure. The majority of the photos are of interiors. The photos are all 400×600 pixels in size.

Download LOL Dataset in Python

Instead of downloading the LOL dataset in Python, you can effortlessly load it in Python via our Deep Lake open source with just one line of code.

Load LOL Dataset Training Subset in Python

				
					import deeplake
ds = deeplake.load('hub://activeloop/lowlight-train')

Load LOL Dataset Validation Subset in Python

				
					import deeplake
ds = deeplake.load('hub://activeloop/lowlight-val')

LOL Dataset Structure

LOL Data Fields

highlight_images: tensor containing images taken at normal light.
lowlight_images: tensor containing images taken at low light.

LOL Data Splits

The LOL dataset training set is composed of 485 samples.
The LOL dataset validation set is composed of 15 samples.

How to use LOL with PyTorch and TensorFlow in Python

Train a model on LOL dataset with PyTorch in Python

Let’s use Deep Lake built-in PyTorch one-line dataloader to connect the data to the compute:

				
					dataloader = ds.pytorch(num_workers=0, batch_size=4, shuffle=False)

Train a model on the LOL dataset with TensorFlow in Python

				
					dataloader = ds.tensorflow()

Additional Information about LOL Dataset

LOL Dataset Description

Homepage: https://daooshee.github.io/BMVC2018website/
Repository: https://github.com/weichen582/RetinexNet
Paper: https://arxiv.org/pdf/2005.02818.pdf

LOL Dataset Curators

Wei Xiong, Ding Liu, Xiaohui Shen, Chen Fang, and Jiebo Luo

LOL Dataset Licensing Information

Deep Lake users may have access to a variety of publicly available datasets. We do not host or distribute these datasets, vouch for their quality or fairness, or claim that you have a license to use the datasets. It is your responsibility to determine whether you have permission to use the datasets under their license. If you’re a dataset owner and do not want your dataset to be included in this library, please get in touch through a GitHub issue. Thank you for your contribution to the ML community!

LOL Dataset Citation Information

				
					@article{wei2018deep,
title={Deep retinex decomposition for low-light enhancement},
author={Wei, Chen and Wang, Wenjing and Yang, Wenhan and Liu, Jiaying},
journal={arXiv preprint arXiv:1808.04560},
year={2018}
}

LOL Dataset FAQs

What is the LOL dataset for Python?

The LOL dataset consists of 500 pairs of low and normal light images, divided into 485 training pairs and 15 test pairs. Low-light images contain noise that occurs during shooting. Most of the images are indoor scenes. All images are 400×600 resolution.

How to download the LOL dataset in Python?

You can load the LOL dataset fast with one line of code using the open-source package Activeloop Deep Lake in Python. See detailed instructions on how to load the LOL dataset training subset and testing subset in Python.

How can I use LOL dataset in PyTorch or TensorFlow?

You can stream the LOL dataset while training a model in PyTorch or TensorFlow with one line of code using the open-source package Activeloop Deep Lake in Python. See detailed instructions on how to train a model on the LOL dataset with PyTorch in Python or train a model on the LOL dataset with TensorFlow in Python.