Croce, V.; Billi, D.; Caroti, G.; Piemonte, A.; De Luca, L.; Véron, P. Comparative Assessment of Neural Radiance Fields and Photogrammetry in Digital Heritage: Impact of Varying Image Conditions on 3D Reconstruction. Remote Sens.2024, 16, 301.
Croce, V.; Billi, D.; Caroti, G.; Piemonte, A.; De Luca, L.; Véron, P. Comparative Assessment of Neural Radiance Fields and Photogrammetry in Digital Heritage: Impact of Varying Image Conditions on 3D Reconstruction. Remote Sens. 2024, 16, 301.
Croce, V.; Billi, D.; Caroti, G.; Piemonte, A.; De Luca, L.; Véron, P. Comparative Assessment of Neural Radiance Fields and Photogrammetry in Digital Heritage: Impact of Varying Image Conditions on 3D Reconstruction. Remote Sens.2024, 16, 301.
Croce, V.; Billi, D.; Caroti, G.; Piemonte, A.; De Luca, L.; Véron, P. Comparative Assessment of Neural Radiance Fields and Photogrammetry in Digital Heritage: Impact of Varying Image Conditions on 3D Reconstruction. Remote Sens. 2024, 16, 301.
Abstract
While Neural Radiance Fields (NeRF) are gaining increasing interest in various domains as innovative methods for novel view synthesis and image-based reconstruction, their potential application in the realm of Cultural Heritage remains unexplored. Purpose of this paper is to assess the effectiveness of applying NeRF to sets of images of digital heritage objects and sites. The study’s findings demonstrate that NeRF could be valuable when used in combination with or as a comparison to other well-established techniques such as photogrammetry, to expand the possibilities of preserving and presenting heritage assets with enhanced visual fidelity and accuracy. Particularly, NeRF show promising results in improving the rendering of translucent and reflective surfaces, objects with homogeneous textures, and elements with intricate details. In addition, we demonstrate that, when considering the same set of input images (with known camera poses), reducing the image quality or the number of images results in significantly less information loss with NeRF compared to photogrammetry. This suggests that NeRF is preferentially suited for scenarios involving sparse information or low-quality photos or videos, which could be especially valuable in risky or challenging situations.
Keywords
NeRF; Cultural Heritage; 3D reconstruction; photogrammetry; 3D surveying
Subject
Engineering, Civil Engineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
