In video denoising, the adjacent frames often provide very useful information, but accurate alignment is needed before such information can be harnassed. In this work, we present a multi-alignment network, which generates multiple flow proposals followed by attention-based averaging. It serves to mimic the non-local mechanism, suppressing noise by averaging multiple observations. Our approach can be applied to various state-of-the-art models that are based on flow estimation. Experiments on a large-scale video dataset demonstrate that our method improves the denoising baseline model by 0.2dB, and further reduces the parameters by 47% with model distillation. Code is available at https://github.com/IndigoPurple/MANet.
@article{arxiv.2202.09704,
title = {MANet: Improving Video Denoising with a Multi-Alignment Network},
author = {Yaping Zhao and Haitian Zheng and Zhongrui Wang and Jiebo Luo and Edmund Y. Lam},
journal= {arXiv preprint arXiv:2202.09704},
year = {2022}
}