期刊
INFORMATION FUSION
卷 93, 期 -, 页码 429-440出版社
ELSEVIER
DOI: 10.1016/j.inffus.2023.01.013
关键词
Computational photography; Video denoising; Low light video; Dark vision; RGB-NIR; Dual-channel network
In this paper, a dual-sensor camera system is proposed to improve video quality under extremely low light conditions by utilizing the correlation between near-infrared (NIR) and RGB spectrum. A compact camera capturing RGB and NIR videos simultaneously is implemented, and a dual-channel multi-frame attention network (DCMAN) is proposed to enhance the quality of low-light RGB and NIR videos. The performance of the camera design and the reconstruction algorithm is validated through experiments on synthetic and real videos.
Videos captured under low light conditions suffer from severe noise. A variety of efforts have been devoted to image/video noise suppression and made large progress. However, in extremely dark scenarios, extensive photon starvation would hamper extracting latent structures buried in noise. Instead, developing an imaging system collecting more photons is a more effective way for high-quality video capture in dark environments. In this paper, we propose to build a dual-sensor camera to additionally collect the photons in NIR wavelength, and make use of the correlation between RGB and near-infrared (NIR) spectrum to perform high-quality reconstruction from noisy dark video pairs. In hardware, we build a compact dual-sensor camera capturing parallax-free RGB and NIR videos simultaneously. Computationally, we propose a dual-channel multi-frame attention network (DCMAN) utilizing spatial-temporal-spectral priors to mutually boost the quality of low-light RGB and NIR videos. In addition, we build a high-quality paired RGB and NIR video dataset, based on which the approach can be applied to different sensors easily by training the DCMAN model with simulated noisy input following a physical-process-based CMOS noise model. Both experiments on synthetic and real videos validate the performance of this compact dual-sensor camera design and the corresponding reconstruction algorithm in dark videography.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据