Back to datasets
Dataset assetOpen Source CommunityObject DetectionLow‑Light Image Processing
Exclusively Dark (ExDark) Image Dataset
To promote research on low‑light object detection and image enhancement, we introduce the Exclusively Dark (ExDark) dataset, which comprises 7,363 images captured under extremely low‑light to dusk conditions (10 different illumination levels) and includes 12 object categories (similar to PASCAL VOC). Images are annotated with both image‑level class labels and localized object bounding boxes.
Source
github
Created
Jun 18, 2019
Updated
Aug 15, 2019
Signals
757 views
Availability
Linked source ready
Overview
Dataset description and usage context
Exclusively Dark (ExDark) Image Dataset
Basic Information
- Release Date: May 29, 2018
- Update Dates:
- June 2, 2019 (low‑light image enhancement code release)
- October 31, 2018 (accepted by CVIU)
Description
- Purpose: Facilitate research on new object detection and image enhancement methods for low‑light environments.
- Content: 7,363 low‑light images covering 10 illumination conditions, with 12 object categories comparable to PASCAL VOC, annotated at the image‑level and with local object bounding boxes.
Structure
- Number of Images: 7,363
- Condition Types: 10
- Object Categories: 12
Code Resources
- Low‑Light Image Enhancement Source Code: Available in the SPIC folder.
Citation
- BibTeX:
@article{Exdark,
title={Getting to Know Low-light Images with The Exclusively Dark Dataset},
author={Loh, Yuen Peng and Chan, Chee Seng},
journal={Computer Vision and Image Understanding},
volume={178},
pages={30-42},
year={2019},
doi={https://doi.org/10.1016/j.cviu.2018.10.010}
}
Need downstream help?
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.