CropNeRF: A Neural Radiance Field-Based Framework for Crop Counting

2026-01-01

arXiv.org

AI Summary

This work proposes a neural radiance field (NeRF)-based 3D instance segmentation method to address the challenge of accurate crop counting in outdoor agricultural settings, where occlusion and high plant density hinder reliable image segmentation. By reconstructing a NeRF from multi-view 2D images and integrating instance masks with visibility and mask consistency scoring mechanisms, the approach establishes a generalizable counting framework that requires no crop-specific parameter tuning. To the best of our knowledge, this is the first study to combine NeRF with 3D instance segmentation for agricultural applications. The method consistently outperforms existing approaches across three datasets—cotton, apple, and pear—and introduces a newly released, publicly available cotton plant dataset to support further research in this domain.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Rigorous crop counting is crucial for effective agricultural management and informed intervention strategies. However, in outdoor field environments, partial occlusions combined with inherent ambiguity in distinguishing clustered crops from individual viewpoints poses an immense challenge for image-based segmentation methods. To address these problems, we introduce a novel crop counting framework designed for exact enumeration via 3D instance segmentation. Our approach utilizes 2D images captured from multiple viewpoints and associates independent instance masks for neural radiance field (NeRF) view synthesis. We introduce crop visibility and mask consistency scores, which are incorporated alongside 3D information from a NeRF model. This results in an effective segmentation of crop instances in 3D and highly-accurate crop counts. Furthermore, our method eliminates the dependence on crop-specific parameter tuning. We validate our framework on three agricultural datasets consisting of cotton bolls, apples, and pears, and demonstrate consistent counting performance despite major variations in crop color, shape, and size. A comparative analysis against the state of the art highlights superior performance on crop counting tasks. Lastly, we contribute a cotton plant dataset to advance further research on this topic.

Problem

• crop counting
• partial occlusion
• clustered crops
• instance ambiguity
• outdoor agricultural environment

Innovation

• Neural Radiance Field
• 3D Instance Segmentation
• Crop Counting
• Multi-view Consistency
• Occlusion Handling