A Compact Dynamic 3D Gaussian Representation for Real-Time Dynamic View Synthesis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Katsumata, Kai, Vo, Duc Minh, Nakayama, Hideki
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916312818647040
author Katsumata, Kai
Vo, Duc Minh
Nakayama, Hideki
author_facet Katsumata, Kai
Vo, Duc Minh
Nakayama, Hideki
contents 3D Gaussian Splatting (3DGS) has shown remarkable success in synthesizing novel views given multiple views of a static scene. Yet, 3DGS faces challenges when applied to dynamic scenes because 3D Gaussian parameters need to be updated per timestep, requiring a large amount of memory and at least a dozen observations per timestep. To address these limitations, we present a compact dynamic 3D Gaussian representation that models positions and rotations as functions of time with a few parameter approximations while keeping other properties of 3DGS including scale, color and opacity invariant. Our method can dramatically reduce memory usage and relax a strict multi-view assumption. In our experiments on monocular and multi-view scenarios, we show that our method not only matches state-of-the-art methods, often linked with slower rendering speeds, in terms of high rendering quality but also significantly surpasses them by achieving a rendering speed of $118$ frames per second (FPS) at a resolution of 1,352$\times$1,014 on a single GPU.
format Preprint
id arxiv_https___arxiv_org_abs_2311_12897
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Compact Dynamic 3D Gaussian Representation for Real-Time Dynamic View Synthesis
Katsumata, Kai
Vo, Duc Minh
Nakayama, Hideki
Graphics
3D Gaussian Splatting (3DGS) has shown remarkable success in synthesizing novel views given multiple views of a static scene. Yet, 3DGS faces challenges when applied to dynamic scenes because 3D Gaussian parameters need to be updated per timestep, requiring a large amount of memory and at least a dozen observations per timestep. To address these limitations, we present a compact dynamic 3D Gaussian representation that models positions and rotations as functions of time with a few parameter approximations while keeping other properties of 3DGS including scale, color and opacity invariant. Our method can dramatically reduce memory usage and relax a strict multi-view assumption. In our experiments on monocular and multi-view scenarios, we show that our method not only matches state-of-the-art methods, often linked with slower rendering speeds, in terms of high rendering quality but also significantly surpasses them by achieving a rendering speed of $118$ frames per second (FPS) at a resolution of 1,352$\times$1,014 on a single GPU.
title A Compact Dynamic 3D Gaussian Representation for Real-Time Dynamic View Synthesis
topic Graphics
url https://arxiv.org/abs/2311.12897