VeGaS: Video Gaussian Splatting

TL;DR

VeGaS introduces a novel Gaussian-based video representation that enables realistic editing and outperforms existing methods in frame reconstruction, addressing limitations of previous neural implicit models.

Contribution

The paper presents VeGaS, a new model using Folded-Gaussian distributions for nonlinear video dynamics, allowing effective editing and improved reconstruction over prior Gaussian splatting approaches.

Findings

Outperforms state-of-the-art in frame reconstruction

Enables realistic video modifications

Models nonlinear dynamics with Folded-Gaussian distributions

Abstract

Implicit Neural Representations (INRs) employ neural networks to approximate discrete data as continuous functions. In the context of video data, such models can be utilized to transform the coordinates of pixel locations along with frame occurrence times (or indices) into RGB color values. Although INRs facilitate effective compression, they are unsuitable for editing purposes. One potential solution is to use a 3D Gaussian Splatting (3DGS) based model, such as the Video Gaussian Representation (VGR), which is capable of encoding video as a multitude of 3D Gaussians and is applicable for numerous video processing operations, including editing. Nevertheless, in this case, the capacity for modification is constrained to a limited set of basic transformations. To address this issue, we introduce the Video Gaussian Splatting (VeGaS) model, which enables realistic modifications of video data. To construct VeGaS, we propose a novel family of Folded-Gaussian distributions designed to capture nonlinear dynamics in a video stream and model consecutive frames by 2D Gaussians obtained as respective conditional distributions. Our experiments demonstrate that VeGaS outperforms state-of-the-art solutions in frame reconstruction tasks and allows realistic modifications of video data. The code is available at: https://github.com/gmum/VeGaS.