Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance Fields

The real-time simulation of large-scale agricultural operations will offer farmers data-driven and physically consistent decision support, facilitated by predictive digital twins. To construct a predictive digital twin, the initial step involves 3D reconstruction of plant geometry. In this paper, a...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiajun Zheng, Xinyi AI, Hao Qin, Jiacheng Rong, Zhiqin Zhang, Yan Yang, Ting Yuan, Wei Li
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10589455/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850109004450103296
author Xiajun Zheng
Xinyi AI
Hao Qin
Jiacheng Rong
Zhiqin Zhang
Yan Yang
Ting Yuan
Wei Li
author_facet Xiajun Zheng
Xinyi AI
Hao Qin
Jiacheng Rong
Zhiqin Zhang
Yan Yang
Ting Yuan
Wei Li
author_sort Xiajun Zheng
collection DOAJ
description The real-time simulation of large-scale agricultural operations will offer farmers data-driven and physically consistent decision support, facilitated by predictive digital twins. To construct a predictive digital twin, the initial step involves 3D reconstruction of plant geometry. In this paper, a high-resolution, accurate 3D reconstruction of tomato plants, Tomato-NeRF, is proposed, which is specially used for three-dimensional reconstruction of tomato plants. Our approach used a modular design to integrate ideas from their research paper into Tomato-NeRF. By using hash encoding to map coordinates to trainable feature vectors, we balance quality, memory usage, and performance in NeRF training. The proposal sampler targets key regions for rendering, and customized loss functions are designed to optimize specific tasks. The effectiveness of our approach is demonstrated by the ability to generate high-resolution geometric models from phone camera data. Comparative results show that Tomato-NeRF has significant advantages over Instant-NGP and MipNeRF in the tomato plant reconstruction task. The data acquisition method is simpler and more efficient than other reconstruction methods, providing a practical solution for real-time agricultural simulations.
format Article
id doaj-art-a02cac5cd66e4f08b97b04b14308718f
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-a02cac5cd66e4f08b97b04b14308718f2025-08-20T02:38:13ZengIEEEIEEE Access2169-35362024-01-011218420618421510.1109/ACCESS.2024.342490810589455Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance FieldsXiajun Zheng0https://orcid.org/0009-0009-7697-8877Xinyi AI1Hao Qin2Jiacheng Rong3Zhiqin Zhang4Yan Yang5Ting Yuan6Wei Li7College of Engineering, China Agricultural University, Beijing, ChinaCollege of Engineering, China Agricultural University, Beijing, ChinaCollege of Engineering, China Agricultural University, Beijing, ChinaCollege of Engineering, China Agricultural University, Beijing, ChinaCollege of Engineering, China Agricultural University, Beijing, ChinaCollege of Engineering, China Agricultural University, Beijing, ChinaCollege of Engineering, China Agricultural University, Beijing, ChinaCollege of Engineering, China Agricultural University, Beijing, ChinaThe real-time simulation of large-scale agricultural operations will offer farmers data-driven and physically consistent decision support, facilitated by predictive digital twins. To construct a predictive digital twin, the initial step involves 3D reconstruction of plant geometry. In this paper, a high-resolution, accurate 3D reconstruction of tomato plants, Tomato-NeRF, is proposed, which is specially used for three-dimensional reconstruction of tomato plants. Our approach used a modular design to integrate ideas from their research paper into Tomato-NeRF. By using hash encoding to map coordinates to trainable feature vectors, we balance quality, memory usage, and performance in NeRF training. The proposal sampler targets key regions for rendering, and customized loss functions are designed to optimize specific tasks. The effectiveness of our approach is demonstrated by the ability to generate high-resolution geometric models from phone camera data. Comparative results show that Tomato-NeRF has significant advantages over Instant-NGP and MipNeRF in the tomato plant reconstruction task. The data acquisition method is simpler and more efficient than other reconstruction methods, providing a practical solution for real-time agricultural simulations.https://ieeexplore.ieee.org/document/10589455/3D reconstructionagricultural automationdeep learningNeRFmulti-view imaging
spellingShingle Xiajun Zheng
Xinyi AI
Hao Qin
Jiacheng Rong
Zhiqin Zhang
Yan Yang
Ting Yuan
Wei Li
Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance Fields
IEEE Access
3D reconstruction
agricultural automation
deep learning
NeRF
multi-view imaging
title Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance Fields
title_full Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance Fields
title_fullStr Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance Fields
title_full_unstemmed Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance Fields
title_short Tomato-Nerf: Advancing Tomato Model Reconstruction With Improved Neural Radiance Fields
title_sort tomato nerf advancing tomato model reconstruction with improved neural radiance fields
topic 3D reconstruction
agricultural automation
deep learning
NeRF
multi-view imaging
url https://ieeexplore.ieee.org/document/10589455/
work_keys_str_mv AT xiajunzheng tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields
AT xinyiai tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields
AT haoqin tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields
AT jiachengrong tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields
AT zhiqinzhang tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields
AT yanyang tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields
AT tingyuan tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields
AT weili tomatonerfadvancingtomatomodelreconstructionwithimprovedneuralradiancefields