Informative Rays Selection for Few-Shot Neural Radiance Fields

Marco Orsingher, Marco Orsingher, Anthony Dell’Eva, Anthony Dell’Eva, Paolo Zani, Paolo Medici, Massimo Bertozzi

2024

Abstract

Neural Radiance Fields (NeRF) have recently emerged as a powerful method for image-based 3D reconstruction, but the lengthy per-scene optimization limits their practical usage, especially in resource-constrained settings. Existing approaches solve this issue by reducing the number of input views and regularizing the learned volumetric representation with either complex losses or additional inputs from other modalities. In this paper, we present KeyNeRF, a simple yet effective method for training NeRF in few-shot scenarios by focusing on key informative rays. Such rays are first selected at camera level by a view selection algorithm that promotes baseline diversity while guaranteeing scene coverage, then at pixel level by sampling from a probability distribution based on local image entropy. Our approach performs favorably against state-of-theart methods, while requiring minimal changes to existing NeRF codebases.

Paper Citation

in Harvard Style

Orsingher M., Dell’Eva A., Zani P., Medici P. and Bertozzi M. (2024). Informative Rays Selection for Few-Shot Neural Radiance Fields. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 253-261. DOI: 10.5220/0012303600003660

in Bibtex Style

@conference{visapp24,
author={Marco Orsingher and Anthony Dell’Eva and Paolo Zani and Paolo Medici and Massimo Bertozzi},
title={Informative Rays Selection for Few-Shot Neural Radiance Fields},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP},
year={2024},
pages={253-261},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012303600003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP
TI - Informative Rays Selection for Few-Shot Neural Radiance Fields
SN - 978-989-758-679-8
AU - Orsingher M.
AU - Dell’Eva A.
AU - Zani P.
AU - Medici P.
AU - Bertozzi M.
PY - 2024
SP - 253
EP - 261
DO - 10.5220/0012303600003660
PB - SciTePress