Cascading Refinement Video Denoising with Uncertainty Adaptivity

TL;DR

This paper presents a cascading refinement approach for video denoising that iteratively improves alignment and image quality, utilizing uncertainty maps to reduce computation and achieve state-of-the-art results.

Contribution

It introduces a novel cascading refinement framework with uncertainty adaptivity for improved video denoising performance.

Findings

Achieved state-of-the-art performance on CRVD dataset

Reduced computational cost by 25% on average

Effectively handles multi-level noise with uncertainty maps

Abstract

Accurate alignment is crucial for video denoising. However, estimating alignment in noisy environments is challenging. This paper introduces a cascading refinement video denoising method that can refine alignment and restore images simultaneously. Better alignment enables restoration of more detailed information in each frame. Furthermore, better image quality leads to better alignment. This method has achieved SOTA performance by a large margin on the CRVD dataset. Simultaneously, aiming to deal with multi-level noise, an uncertainty map was created after each iteration. Because of this, redundant computation on the easily restored videos was avoided. By applying this method, the entire computation was reduced by 25% on average.