Deep Change Monitoring: A Hyperbolic Representative Learning Framework and a Dataset for Long-term Fine-grained Tree Change Detection

2025-03-01

AI Summary

Low image resolution and high annotation costs hinder fine-grained tree change monitoring in long-term ecological observation. To address this, we introduce UAVTC—the first high-resolution, time-series UAV dataset specifically designed for long-term tree change analysis—and propose the Hyperbolic Space Siamese Network (HSN), an ecologically constrained deep learning framework. HSN innovatively incorporates botanical priors into a fine-grained biologically aware annotation scheme and pioneers the use of hyperbolic geometry in remote sensing time-series change detection, enabling compact representation of hierarchical and nonlinear physiological tree dynamics. Evaluated on UAVTC, HSN achieves significant gains in fine-grained change identification accuracy. Moreover, when transferred to facial anti-spoofing—a domain with distinct data distribution—HSN maintains superior performance, demonstrating strong cross-domain generalization and broad representational utility.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

In environmental protection, tree monitoring plays an essential role in maintaining and improving ecosystem health. However, precise monitoring is challenging because existing datasets fail to capture continuous fine-grained changes in trees due to low-resolution images and high acquisition costs. In this paper, we introduce UAVTC, a large-scale, long-term, high-resolution dataset collected using UAVs equipped with cameras, specifically designed to detect individual Tree Changes (TCs). UAVTC includes rich annotations and statistics based on biological knowledge, offering a fine-grained view for tree monitoring. To address environmental influences and effectively model the hierarchical diversity of physiological TCs, we propose a novel Hyperbolic Siamese Network (HSN) for TC detection, enabling compact and hierarchical representations of dynamic tree changes. Extensive experiments show that HSN can effectively capture complex hierarchical changes and provide a robust solution for fine-grained TC detection. In addition, HSN generalizes well to cross-domain face anti-spoofing task, highlighting its broader significance in AI. We believe our work, combining ecological insights and interdisciplinary expertise, will benefit the community by offering a new benchmark and innovative AI technologies.

Problem

• Detects fine-grained tree changes over time using high-resolution UAV imagery.
• Addresses challenges in tree monitoring due to low-resolution data and high costs.
• Proposes a Hyperbolic Siamese Network for hierarchical representation of tree changes.

Innovation

• UAV-based high-resolution dataset for tree monitoring
• Hyperbolic Siamese Network for hierarchical change detection
• Cross-domain application in face anti-spoofing