Relaxing Accurate Initialization Constraint for 3D Gaussian Splatting

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jung, Jaewoo, Han, Jisang, An, Honggyu, Kang, Jiwon, Park, Seonghoon, Kim, Seungryong
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913366445916160
author Jung, Jaewoo
Han, Jisang
An, Honggyu
Kang, Jiwon
Park, Seonghoon
Kim, Seungryong
author_facet Jung, Jaewoo
Han, Jisang
An, Honggyu
Kang, Jiwon
Park, Seonghoon
Kim, Seungryong
contents 3D Gaussian splatting (3DGS) has recently demonstrated impressive capabilities in real-time novel view synthesis and 3D reconstruction. However, 3DGS heavily depends on the accurate initialization derived from Structure-from-Motion (SfM) methods. When the quality of the initial point cloud deteriorates, such as in the presence of noise or when using randomly initialized point cloud, 3DGS often undergoes large performance drops. To address this limitation, we propose a novel optimization strategy dubbed RAIN-GS (Relaing Accurate Initialization Constraint for 3D Gaussian Splatting). Our approach is based on an in-depth analysis of the original 3DGS optimization scheme and the analysis of the SfM initialization in the frequency domain. Leveraging simple modifications based on our analyses, RAIN-GS successfully trains 3D Gaussians from sub-optimal point cloud (e.g., randomly initialized point cloud), effectively relaxing the need for accurate initialization. We demonstrate the efficacy of our strategy through quantitative and qualitative comparisons on multiple datasets, where RAIN-GS trained with random point cloud achieves performance on-par with or even better than 3DGS trained with accurate SfM point cloud. Our project page and code can be found at https://ku-cvlab.github.io/RAIN-GS.
format Preprint
id arxiv_https___arxiv_org_abs_2403_09413
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Relaxing Accurate Initialization Constraint for 3D Gaussian Splatting
Jung, Jaewoo
Han, Jisang
An, Honggyu
Kang, Jiwon
Park, Seonghoon
Kim, Seungryong
Computer Vision and Pattern Recognition
3D Gaussian splatting (3DGS) has recently demonstrated impressive capabilities in real-time novel view synthesis and 3D reconstruction. However, 3DGS heavily depends on the accurate initialization derived from Structure-from-Motion (SfM) methods. When the quality of the initial point cloud deteriorates, such as in the presence of noise or when using randomly initialized point cloud, 3DGS often undergoes large performance drops. To address this limitation, we propose a novel optimization strategy dubbed RAIN-GS (Relaing Accurate Initialization Constraint for 3D Gaussian Splatting). Our approach is based on an in-depth analysis of the original 3DGS optimization scheme and the analysis of the SfM initialization in the frequency domain. Leveraging simple modifications based on our analyses, RAIN-GS successfully trains 3D Gaussians from sub-optimal point cloud (e.g., randomly initialized point cloud), effectively relaxing the need for accurate initialization. We demonstrate the efficacy of our strategy through quantitative and qualitative comparisons on multiple datasets, where RAIN-GS trained with random point cloud achieves performance on-par with or even better than 3DGS trained with accurate SfM point cloud. Our project page and code can be found at https://ku-cvlab.github.io/RAIN-GS.
title Relaxing Accurate Initialization Constraint for 3D Gaussian Splatting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.09413