OmnimatteRF: Robust Omnimatte with 3D Background Modeling

TL;DR

OmnimatteRF introduces a novel video matting approach that combines 2D foreground layers with a 3D background model, enabling more accurate scene reconstruction in real-world videos.

Contribution

The paper presents OmnimatteRF, a new method that integrates 3D background modeling with dynamic 2D foreground layers for improved video matting.

Findings

Better scene reconstruction quality on various videos

Effective separation of foreground and complex backgrounds

Robust handling of real-world video complexities

Abstract

Video matting has broad applications, from adding interesting effects to casually captured movies to assisting video production professionals. Matting with associated effects such as shadows and reflections has also attracted increasing research activity, and methods like Omnimatte have been proposed to separate dynamic foreground objects of interest into their own layers. However, prior works represent video backgrounds as 2D image layers, limiting their capacity to express more complicated scenes, thus hindering application to real-world videos. In this paper, we propose a novel video matting method, OmnimatteRF, that combines dynamic 2D foreground layers and a 3D background model. The 2D layers preserve the details of the subjects, while the 3D background robustly reconstructs scenes in real-world videos. Extensive experiments demonstrate that our method reconstructs scenes with better quality on various videos.