Videos account for a significant portion of internet traffic, and the presence of noise, whether from compression algorithms, low light, sensor imperfections, deteriorates the video quality. Ambient noise can also significantly diminish the visual quality. Traditional CNN-based video denoising methods rely on convolutional filters with fixed sizes and receptive fields, excelling at capturing local patterns and short-range dependencies. However, CNNs often struggle to handle long-term dependencies or relationships that extend over larger spatial and temporal scales. These are vital for accurately removing noise while preserving essential video details, textures, and structures. To address this limitation, we propose a novel approach, using UNet architecture, which combines the strengths of convolutional neural networks (CNNs) and graph neural networks (GNNs) for local and global information and dependency preservation. In this approach, CNN is followed by transformer attention for sparse graph formation for CNN. The spatiotemporal patches act as nodes, and the similarity between them represent edges. By integrating CNNs for local feature extraction followed by transformer attention and GNN for video denoising first time, for long-term spatio-temporal relationships, improves the ability to accurately model noise, preserve fine details and subsequently denoise videos more accurately. The strong ablation studies prove the effectiveness of the different modules, patch sizes on four different noise types. The proposed method outperformed most of the SOTA video denoising algorithms in terms of both PSNR and SSIM, at moderate computational cost, apart from the Video Restoration Transformer(VRT).

Quality video services have already gained high technical and commercial importance. The published work so far in this domain proposed mathematically and computationally complex algorithms, followed by the recent training-greedy deep learning-based denoising algorithms. This work proposes a video-denoising algorithm based on multiple UNet networks. The proposed video-denoising algorithm uses multiple encoder-decoder networks for video noise residual frame estimation, un-like the single encoder-decoder used by the published denoising algorithms. Using multiple skip connection UNets, we increase the residual noise modeling accuracy while restricting the signal features, which helps to improve denoising performance. The proposed network is trained end-to-end without motion compensation to reduce its complexity. The proposed network outperforms all the video denoising algorithms in terms of SSIM metric while it yields comparable performance in terms of PSNR.