KIT | KIT-Bibliothek | Impressum | Datenschutz

Leveraging Neural Radiance Fields for Large-Scale 3D Reconstruction from Aerial Imagery

Hermann, Max 1; Kwak, Hyovin; Ruf, Boitumelo 1; Weinmann, Martin 1
1 Institut für Photogrammetrie und Fernerkundung (IPF), Karlsruher Institut für Technologie (KIT)

Abstract:

Since conventional photogrammetric approaches struggle with with low-texture, reflective, and transparent regions, this study explores the application of Neural Radiance Fields (NeRFs) for large-scale 3D reconstruction of outdoor scenes, since NeRF-based methods have recently shown very impressive results in these areas. We evaluate three approaches: Mega-NeRF, Block-NeRF, and Direct Voxel Grid Optimization, focusing on their accuracy and completeness compared to ground
truth point clouds. In addition, we analyze the effects of using multiple sub-modules, estimating the visibility by an additional neural network and varying the density threshold for the extraction of the point cloud. For performance valuation, we use benchmark datasets that correspond to the setting off standard flight campaigns and therefore typically have nadir camera perspective and relatively
little image overlap, which can be challenging for NeRF-based approaches that are typically trained
with significantly more images and varying camera angles. We show that despite lower quality
compared to classic photogrammetric approaches, NeRF-based reconstructions provide visually
... mehr

Zugehörige Institution(en) am KIT Institut für Photogrammetrie und Fernerkundung (IPF)
Publikationstyp Zeitschriftenaufsatz
Publikationsdatum 12.12.2024
Sprache Englisch
Identifikator ISSN: 2072-4292
KITopen-ID: 1000178502
Erschienen in Remote Sensing
Verlag MDPI
Band 16
Heft 24
Seiten Art.-Nr.: 4655
Nachgewiesen in Scopus
Web of Science
Dimensions

Verlagsausgabe §
DOI: 10.5445/IR/1000178502
Veröffentlicht am 29.01.2025
Seitenaufrufe: 17
seit 29.01.2025
Downloads: 10
seit 02.02.2025
Cover der Publikation
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page