Gaussian Splatting for Efficient Satellite Image Photogrammetry

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Aira, Luca Savant, Facciolo, Gabriele, Ehret, Thibaud
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915210816651264
author Aira, Luca Savant
Facciolo, Gabriele
Ehret, Thibaud
author_facet Aira, Luca Savant
Facciolo, Gabriele
Ehret, Thibaud
contents Recently, Gaussian splatting has emerged as a strong alternative to NeRF, demonstrating impressive 3D modeling capabilities while requiring only a fraction of the training and rendering time. In this paper, we show how the standard Gaussian splatting framework can be adapted for remote sensing, retaining its high efficiency. This enables us to achieve state-of-the-art performance in just a few minutes, compared to the day-long optimization required by the best-performing NeRF-based Earth observation methods. The proposed framework incorporates remote-sensing improvements from EO-NeRF, such as radiometric correction and shadow modeling, while introducing novel components, including sparsity, view consistency, and opacity regularizations.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13047
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Gaussian Splatting for Efficient Satellite Image Photogrammetry
Aira, Luca Savant
Facciolo, Gabriele
Ehret, Thibaud
Computer Vision and Pattern Recognition
Recently, Gaussian splatting has emerged as a strong alternative to NeRF, demonstrating impressive 3D modeling capabilities while requiring only a fraction of the training and rendering time. In this paper, we show how the standard Gaussian splatting framework can be adapted for remote sensing, retaining its high efficiency. This enables us to achieve state-of-the-art performance in just a few minutes, compared to the day-long optimization required by the best-performing NeRF-based Earth observation methods. The proposed framework incorporates remote-sensing improvements from EO-NeRF, such as radiometric correction and shadow modeling, while introducing novel components, including sparsity, view consistency, and opacity regularizations.
title Gaussian Splatting for Efficient Satellite Image Photogrammetry
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.13047