EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

High-fidelity wheat plant reconstruction using 3D Gaussian splatting and neural radiance fields

Lewis A G Stuart iD, Darren M Wells iD, Jonathan A Atkinson iD, Simon Castle-Green iD, Jack Walker, Michael P Pound iD

GigaScience · 06.01.2025 · 14

DOI: 10.1093/gigascience/giaf022 ↗

Kurz gefasst

Fachartikel in der Zeitschrift GigaScience, online erschienen am 26.03.2025. Verfasst von Lewis A G Stuart und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science, University of Nottingham, Nottingham, NG8 1BB, UK. Hauptthemen laut MEDLINE: Triticum, Imaging, Three-Dimensional, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 37-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Nottingham.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The reconstruction of 3-dimensional (3D) plant models can offer advantages over traditional 2-dimensional approaches by more accurately capturing the complex structure and characteristics of different crops. Conventional 3D reconstruction techniques often produce sparse or noisy representations of plants using software or are expensive to capture in hardware. Recently, view synthesis models have been developed that can generate detailed 3D scenes, and even 3D models, from only RGB images and camera poses. These models offer unparalleled accuracy but are currently data hungry, requiring large numbers of views with very accurate camera calibration.

Results

In this study, we present a view synthesis dataset comprising 20 individual wheat plants captured across 6 different time frames over a 15-week growth period. We develop a camera capture system using 2 robotic arms combined with a turntable, controlled by a re-deployable and flexible image capture framework. We trained each plant instance using two recent view synthesis models: 3D Gaussian splatting (3DGS) and neural radiance fields (NeRF). Our results show that both 3DGS and NeRF produce high-fidelity reconstructed images of a plant subject from views not captured in the initial training sets. We also show that these approaches can be used to generate accurate 3D representations of these plants as point clouds, with 0.74-mm and 1.43-mm average accuracy compared with a handheld scanner for 3DGS and NeRF, respectively.

Conclusion

We believe that these new methods will be transformative in the field of 3D plant phenotyping, plant reconstruction, and active vision. To further this cause, we release all robot configuration and control software, alongside our extensive multiview dataset. We also release all scripts necessary to train both 3DGS and NeRF, all trained models data, and final 3D point cloud representations. Our dataset can be accessed via https://plantimages.nottingham.ac.uk/ or https://https://doi.org/10.5524/102661. Our software can be accessed via https://github.com/Lewis-Stuart-11/3D-Plant-View-Synthesis.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

TriticumImaging, Three-DimensionalImage Processing, Computer-AssistedNormal DistributionSoftwareAlgorithms

Stichwörter der Autor:innen

3D Gaussian splatting3D reconstruction3DGSNeRFdigital twinimagingmachine learningneural radiance fieldsphenotypingroboticsview synthesis

Autor:innen und Institutionen

  • School of Computer Science, University of Nottingham, Nottingham, NG8 1BB, UK.
  • School of Biosciences, University of Nottingham, Nottingham, LE12 5RD, UK.

ORCID-Profile: Lewis A G Stuart, Darren M Wells, Jonathan A Atkinson, Simon Castle-Green, Michael P Pound

Förderung

  • University of Nottingham

Zitieren

APA

Stuart, L. A. G., Wells, D. M., Atkinson, J. A., Castle-Green, S., Walker, J., & Pound, M. P. (2025). High-fidelity wheat plant reconstruction using 3D Gaussian splatting and neural radiance fields. GigaScience, 14. https://doi.org/10.1093/gigascience/giaf022

Vancouver

Stuart LAG, Wells DM, Atkinson JA, Castle-Green S, Walker J, Pound MP. High-fidelity wheat plant reconstruction using 3D Gaussian splatting and neural radiance fields. GigaScience. 2025;14. doi:10.1093/gigascience/giaf022

BibTeX
@article{Stuart2025,
  author = {Stuart, Lewis A G and Wells, Darren M and Atkinson, Jonathan A and Castle-Green, Simon and Walker, Jack and Pound, Michael P},
  title = {High-fidelity wheat plant reconstruction using 3D Gaussian splatting and neural radiance fields},
  journal = {GigaScience},
  year = {2025},
  volume = {14},
  doi = {10.1093/gigascience/giaf022}
}
Kostenlos starten →