Model-Blind Video Denoising via Frame-to-Frame Training
Abstract
Modeling the processing chain that has produced a video is a difficult reverse engineering task, even when the camera is available. This makes model based video processing a still more complex task. In this paper we propose a fully blind video denoising method, with two versions off-line and on-line. This is achieved by fine-tuning a pre-trained AWGN denoising network to the video with a novel frame-to-frame training strategy. Our denoiser can be used without knowledge of the origin of the video or burst and the post-processing steps applied from the camera sensor. The on-line process only requires a couple of frames before achieving visually pleasing results for a wide range of perturbations. It nonetheless reaches state-of-the-art performance for standard Gaussian noise, and can be used off-line with still better performance.
Cite
Text
Ehret et al. "Model-Blind Video Denoising via Frame-to-Frame Training." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019. doi:10.1109/CVPR.2019.01163Markdown
[Ehret et al. "Model-Blind Video Denoising via Frame-to-Frame Training." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019.](https://mlanthology.org/cvpr/2019/ehret2019cvpr-modelblind/) doi:10.1109/CVPR.2019.01163BibTeX
@inproceedings{ehret2019cvpr-modelblind,
title = {{Model-Blind Video Denoising via Frame-to-Frame Training}},
author = {Ehret, Thibaud and Davy, Axel and Morel, Jean-Michel and Facciolo, Gabriele and Arias, Pablo},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year = {2019},
doi = {10.1109/CVPR.2019.01163},
url = {https://mlanthology.org/cvpr/2019/ehret2019cvpr-modelblind/}
}