A qualitative investigation of optical flow algorithms for video denoising

TL;DR

This paper qualitatively evaluates the performance of various optical flow algorithms, including classic and deep learning methods, within a state-of-the-art video denoising framework on challenging real-world content.

Contribution

It provides a comparative analysis of optical flow algorithms' effectiveness in video denoising under realistic and challenging conditions.

Findings

Deep learning optical flow algorithms outperform classic methods in noisy scenarios.

Optical flow quality significantly impacts denoising performance.

Challenging content reveals limitations of current optical flow algorithms.

Abstract

A good optical flow estimation is crucial in many video analysis and restoration algorithms employed in application fields like media industry, industrial inspection and automotive. In this work, we investigate how well optical flow algorithms perform qualitatively when integrated into a state of the art video denoising algorithm. Both classic optical flow algorithms (e.g. TV-L1) as well as recent deep learning based algorithm (like RAFT or BMBC) will be taken into account. For the qualitative investigation, we will employ realistic content with challenging characteristic (noisy content, large motion etc.) instead of the standard images used in most publications.