3DAeroRelief: The first 3D Benchmark UAV Dataset for Post-Disaster Assessment

2025-09-14

AI Summary

Existing 3D semantic segmentation benchmarks predominantly target urban or indoor scenes, lacking outdoor 3D data specifically designed for post-disaster assessment. To address this gap, we introduce Hurricane3D—the first dedicated 3D semantic segmentation benchmark for hurricane-induced structural damage evaluation. It comprises large-scale, high-density outdoor point clouds reconstructed via Structure-from-Motion (SfM) and Multi-View Stereo (MVS) from multi-view aerial imagery captured by low-cost UAVs in real disaster zones. We propose a fine-grained annotation paradigm that projects 2D pixel-level damage labels onto 3D point clouds, covering detailed damage categories including collapse, tilting, and cracking. Hurricane3D fills a critical void in disaster-scene 3D vision benchmarks and enables scalable 3D understanding of post-disaster environments. Extensive evaluation of state-of-the-art 3D segmentation models reveals key challenges: severe occlusion, geometric degradation, and extreme class imbalance. This benchmark establishes a new foundation for 3D perception in emergency response.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Reasoning under Uncertainty: Uncertainty Representations

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
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Timely assessment of structural damage is critical for disaster response and recovery. However, most prior work in natural disaster analysis relies on 2D imagery, which lacks depth, suffers from occlusions, and provides limited spatial context. 3D semantic segmentation offers a richer alternative, but existing 3D benchmarks focus mainly on urban or indoor scenes, with little attention to disaster-affected areas. To address this gap, we present 3DAeroRelief--the first 3D benchmark dataset specifically designed for post-disaster assessment. Collected using low-cost unmanned aerial vehicles (UAVs) over hurricane-damaged regions, the dataset features dense 3D point clouds reconstructed via Structure-from-Motion and Multi-View Stereo techniques. Semantic annotations were produced through manual 2D labeling and projected into 3D space. Unlike existing datasets, 3DAeroRelief captures 3D large-scale outdoor environments with fine-grained structural damage in real-world disaster contexts. UAVs enable affordable, flexible, and safe data collection in hazardous areas, making them particularly well-suited for emergency scenarios. To demonstrate the utility of 3DAeroRelief, we evaluate several state-of-the-art 3D segmentation models on the dataset to highlight both the challenges and opportunities of 3D scene understanding in disaster response. Our dataset serves as a valuable resource for advancing robust 3D vision systems in real-world applications for post-disaster scenarios.

Problem

• Lack of 3D benchmark datasets for post-disaster damage assessment
• Existing 3D datasets focus on urban scenes not disaster areas
• 2D imagery lacks depth and spatial context for damage analysis

Innovation

• UAV-collected 3D point clouds
• Structure-from-Motion reconstruction
• 3D semantic segmentation annotations