# DVDnet: A Fast Network for Deep Video Denoising

## TL;DR

DVDnet introduces a fast, neural network-based video denoising method that outperforms traditional patch-based approaches in quality and speed, with a small memory footprint and versatility across noise levels.

## Contribution

The paper presents a novel neural network architecture for video denoising that surpasses existing patch-based methods in performance and efficiency.

## Key findings

- Outperforms state-of-the-art patch-based denoising methods
- Operates with significantly lower computational times
- Handles a wide range of noise levels with a single model

## Abstract

In this paper, we propose a state-of-the-art video denoising algorithm based on a convolutional neural network architecture. Previous neural network based approaches to video denoising have been unsuccessful as their performance cannot compete with the performance of patch-based methods. However, our approach outperforms other patch-based competitors with significantly lower computing times. In contrast to other existing neural network denoisers, our algorithm exhibits several desirable properties such as a small memory footprint, and the ability to handle a wide range of noise levels with a single network model. The combination between its denoising performance and lower computational load makes this algorithm attractive for practical denoising applications. We compare our method with different state-of-art algorithms, both visually and with respect to objective quality metrics. The experiments show that our algorithm compares favorably to other state-of-art methods. Video examples, code and models are publicly available at \url{https://github.com/m-tassano/dvdnet}.