Enregistré dans:
Détails bibliographiques
Auteurs principaux: Xiong, Tianyi, Wu, Jiayi, He, Botao, Fermuller, Cornelia, Aloimonos, Yiannis, Huang, Heng, Metzler, Christopher A.
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2406.02972
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910647654023168
author Xiong, Tianyi
Wu, Jiayi
He, Botao
Fermuller, Cornelia
Aloimonos, Yiannis
Huang, Heng
Metzler, Christopher A.
author_facet Xiong, Tianyi
Wu, Jiayi
He, Botao
Fermuller, Cornelia
Aloimonos, Yiannis
Huang, Heng
Metzler, Christopher A.
contents By combining differentiable rendering with explicit point-based scene representations, 3D Gaussian Splatting (3DGS) has demonstrated breakthrough 3D reconstruction capabilities. However, to date 3DGS has had limited impact on robotics, where high-speed egomotion is pervasive: Egomotion introduces motion blur and leads to artifacts in existing frame-based 3DGS reconstruction methods. To address this challenge, we introduce Event3DGS, an {\em event-based} 3DGS framework. By exploiting the exceptional temporal resolution of event cameras, Event3GDS can reconstruct high-fidelity 3D structure and appearance under high-speed egomotion. Extensive experiments on multiple synthetic and real-world datasets demonstrate the superiority of Event3DGS compared with existing event-based dense 3D scene reconstruction frameworks; Event3DGS substantially improves reconstruction quality (+3dB) while reducing computational costs by 95\%. Our framework also allows one to incorporate a few motion-blurred frame-based measurements into the reconstruction process to further improve appearance fidelity without loss of structural accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2406_02972
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Event3DGS: Event-Based 3D Gaussian Splatting for High-Speed Robot Egomotion
Xiong, Tianyi
Wu, Jiayi
He, Botao
Fermuller, Cornelia
Aloimonos, Yiannis
Huang, Heng
Metzler, Christopher A.
Computer Vision and Pattern Recognition
By combining differentiable rendering with explicit point-based scene representations, 3D Gaussian Splatting (3DGS) has demonstrated breakthrough 3D reconstruction capabilities. However, to date 3DGS has had limited impact on robotics, where high-speed egomotion is pervasive: Egomotion introduces motion blur and leads to artifacts in existing frame-based 3DGS reconstruction methods. To address this challenge, we introduce Event3DGS, an {\em event-based} 3DGS framework. By exploiting the exceptional temporal resolution of event cameras, Event3GDS can reconstruct high-fidelity 3D structure and appearance under high-speed egomotion. Extensive experiments on multiple synthetic and real-world datasets demonstrate the superiority of Event3DGS compared with existing event-based dense 3D scene reconstruction frameworks; Event3DGS substantially improves reconstruction quality (+3dB) while reducing computational costs by 95\%. Our framework also allows one to incorporate a few motion-blurred frame-based measurements into the reconstruction process to further improve appearance fidelity without loss of structural accuracy.
title Event3DGS: Event-Based 3D Gaussian Splatting for High-Speed Robot Egomotion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.02972