PlantTraitNet: An Uncertainty-Aware Multimodal Framework for Global-Scale Plant Trait Inference from Citizen Science Data

2025-11-10

AI Summary

Traditional plant trait mapping (e.g., height, leaf area) relies on sparse and costly field measurements, limiting scalability for global ecosystem modeling. To address this, we propose the first uncertainty-aware multi-task, multimodal deep learning framework that leverages over 50 million geotagged citizen-science plant images worldwide. By jointly encoding visual and geospatial features, it enables weakly supervised remote sensing retrieval of plant traits without pixel-level annotations—using only image-level geographic coordinates and coarse metadata. Key contributions include: (i) the first integration of predictive uncertainty quantification into multimodal joint training; (ii) a fully weakly supervised paradigm eliminating the need for labor-intensive fine-grained labeling; and (iii) generation of globally continuous, high-resolution trait distribution maps. Our framework achieves statistically significant improvements in both accuracy and spatial coverage across all six evaluated traits, outperforming state-of-the-art products, and is rigorously validated against independent field survey data.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Security and Privacy: Data transparency and provenance

Abstract

Global plant maps of plant traits, such as leaf nitrogen or plant height, are essential for understanding ecosystem processes, including the carbon and energy cycles of the Earth system. However, existing trait maps remain limited by the high cost and sparse geographic coverage of field-based measurements. Citizen science initiatives offer a largely untapped resource to overcome these limitations, with over 50 million geotagged plant photographs worldwide capturing valuable visual information on plant morphology and physiology. In this study, we introduce PlantTraitNet, a multi-modal, multi-task uncertainty-aware deep learning framework that predictsfour key plant traits (plant height, leaf area, specific leaf area, and nitrogen content) from citizen science photos using weak supervision. By aggregating individual trait predictions across space, we generate global maps of trait distributions. We validate these maps against independent vegetation survey data (sPlotOpen) and benchmark them against leading global trait products. Our results show that PlantTraitNet consistently outperforms existing trait maps across all evaluated traits, demonstrating that citizen science imagery, when integrated with computer vision and geospatial AI, enables not only scalable but also more accurate global trait mapping. This approach offers a powerful new pathway for ecological research and Earth system modeling.

Problem

• Predicting plant traits from citizen science photos using deep learning
• Generating global trait maps by aggregating spatial predictions
• Validating trait maps against independent ecological survey data

Innovation

• Multimodal deep learning framework for plant trait inference
• Uses citizen science photos with weak supervision
• Generates global maps with uncertainty-aware predictions