Dancing in the Dark: A Benchmark towards General Low-light Video Enhancement
TL;DR: This paper introduces a high-quality low-light video dataset (DID) and a Retinex-based method called Light Adjustable Network (LAN) for general low-light video enhancement. The dataset features dynamic videos with multiple exposures and cameras, while LAN iteratively refines illumination for adaptive enhancement.
Abstract
Low-light video enhancement is a challenging task with broad applications. However, current research in this area is limited by the lack of high-quality benchmark datasets. To address this issue, we design a camera system and collect a high-quality low-light video dataset with multiple exposures and cameras. Our dataset provides dynamic video pairs with pronounced camera motion and strict spatial alignment. To achieve general low-light video enhancement, we also propose a novel Retinex-based method named Light Adjustable Network (LAN). LAN iteratively refines the illumination and adaptively adjusts it under varying lighting conditions, leading to visually appealing results even in diverse real-world scenarios. The extensive experiments demonstrate the superiority of our low-light video dataset and enhancement method. Our dataset is available at https://github.com/ciki000/DID.
Optical System to Collect Dataset
@inproceedings{fuDancingDarkICCV2023,
author={Fu, Huiyuan and Zheng, Wenkai and Wang, Xicong and Wang, Jiaxuan and Zhang, Heng and Ma, Huadong},
booktitle={IEEE/CVF International Conference on Computer Vision (ICCV)},
title={Dancing in the Dark: A Benchmark towards General Low-light Video Enhancement},
year={2023},
pages={12831-12840}
}