OmniPlantSeg: Species Agnostic 3D Point Cloud Organ Segmentation for High-Resolution Plant Phenotyping Across Modalities

2025-09-25

AI Summary

Existing plant organ point cloud segmentation methods suffer from species specificity, sensor dependency, and mandatory downsampling—leading to poor generalizability and severe resolution loss. To address these limitations, we propose KDSS (K-Distance Subsampling), a novel subsampling algorithm that operates directly on raw high-resolution point clouds without preprocessing or aggressive downsampling. KDSS enables cross-species applicability (e.g., tomato, maize, Arabidopsis thaliana) and robustness across multimodal acquisition modalities—including photogrammetry, laser triangulation, and LiDAR. Integrated seamlessly with state-of-the-art point cloud segmentation architectures (e.g., PointNet++, KPConv), KDSS preserves full spatial fidelity while significantly improving segmentation accuracy. Extensive experiments demonstrate consistent superiority over conventional downsampling strategies across diverse species and sensing modalities. This work establishes a new paradigm for generalizable, high-fidelity plant phenotyping.

Technology Category

• Computer Vision: Segmentation
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Data Mining & Knowledge Management: Data Compression

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Accurate point cloud segmentation for plant organs is crucial for 3D plant phenotyping. Existing solutions are designed problem-specific with a focus on certain plant species or specified sensor-modalities for data acquisition. Furthermore, it is common to use extensive pre-processing and down-sample the plant point clouds to meet hardware or neural network input size requirements. We propose a simple, yet effective algorithm KDSS for sub-sampling of biological point clouds that is agnostic to sensor data and plant species. The main benefit of this approach is that we do not need to down-sample our input data and thus, enable segmentation of the full-resolution point cloud. Combining KD-SS with current state-of-the-art segmentation models shows satisfying results evaluated on different modalities such as photogrammetry, laser triangulation and LiDAR for various plant species. We propose KD-SS as lightweight resolution-retaining alternative to intensive pre-processing and down-sampling methods for plant organ segmentation regardless of used species and sensor modality.

Problem

• Accurate 3D point cloud segmentation for plant organs across species
• Eliminating intensive preprocessing and downsampling of high-resolution data
• Creating sensor-agnostic method for various acquisition modalities

Innovation

• KDSS sub-sampling algorithm for biological point clouds
• Segmentation of full-resolution point cloud without down-sampling
• Lightweight method agnostic to plant species and sensor modality