paper-with-me

홈 › Papers

Deep Parametric 3D Filters for Joint Video Denoising and Illumination Enhancement in Video Super Resolution

2022-07-05 · Xiaogang Xu, RuiXing Wang, Chi-Wing Fu, Jiaya Jia

Despite the quality improvement brought by the recent methods, video super-resolution (SR) is still very challenging, especially for videos that are low-light and noisy. The current best solution is to subsequently employ best models of video SR, denoising, and illumination enhancement, but doing so often lowers the image quality, due to the inconsistency between the models. This paper presents a new parametric representation called the Deep Parametric 3D Filters (DP3DF), which incorporates local spatiotemporal information to enable simultaneous denoising, illumination enhancement, and SR efficiently in a single encoder-and-decoder network. Also, a dynamic residual frame is jointly learned with the DP3DF via a shared backbone to further boost the SR quality. We performed extensive experiments, including a large-scale user study, to show our method's effectiveness. Our method consistently surpasses the best state-of-the-art methods on all the challenging real datasets with top PSNR and user ratings, yet having a very fast run time.

📄 PDF Abstract BibTeX arXiv:2207.01797

Code (1)

xiaogang00/dp3df 공식 구현 pytorch

Tasks

DecoderDenoisingSuper-ResolutionVideo DenoisingVideo Super-Resolution

Similar Papers 제목 키워드 기반

ZERO-IG: Zero-Shot Illumination-Guided Joint Denoising and Adaptive Enhancement for Low-Light Images

2024-01-01 · CVPR 2024 1 · Yiqi Shi, Duo Liu, Liguo Zhang, Ye Tian 외

This paper presents a novel zero-shot method for jointly denoising and enhancing real-word low-light images. The proposed method is independent of training data and noise distribution. Guided by illumination we integ…

Denoising

Dancing under the stars: video denoising in starlight

2022-04-08 · CVPR 2022 1 · Kristina Monakhova, Stephan R. Richter, Laura Waller, Vladlen Koltun

Imaging in low light is extremely challenging due to low photon counts. Using sensitive CMOS cameras, it is currently possible to take videos at night under moonlight (0.05-0.3 lux illumination). In this paper, we demons…

DenoisingImage DenoisingVideo Denoising

Progressive Joint Low-light Enhancement and Noise Removal for Raw Images

2021-06-28 · Yucheng Lu, Seung-Won Jung

Low-light imaging on mobile devices is typically challenging due to insufficient incident light coming through the relatively small aperture, resulting in a low signal-to-noise ratio. Most of the previous works on low-li…

Denoising

Exploiting Temporal Attention Features for Effective Denoising in Videos

2020-08-05 · Aryansh Omray, Samyak Jain, Utsav Krishnan, Pratik Chattopadhyay

Video Denoising is one of the fundamental tasks of any videoprocessing pipeline. It is different from image denoising due to the tem-poral aspects of video frames, and any image denoising approach appliedto videos will r…

DecoderDenoisingImage DenoisingSSIM+1

Joint Enhancement and Denoising Method via Sequential Decomposition

2018-04-23 · Xutong Ren, Mading Li, Wen-Huang Cheng, Jiaying Liu

Many low-light enhancement methods ignore intensive noise in original images. As a result, they often simultaneously enhance the noise as well. Furthermore, extra denoising procedures adopted by most methods ruin the det…

Denoising