Damage-TriageFormer: A Foundation-Model Framework for Typology-Based Building Damage Assessment from Mono-Temporal Imagery
作者: Yiming Xiao, Yu-Hsuan Ho, Sanjay Thasma, Junwei Ma, Ali Mostafavi
分类: cs.CV
发布日期: 2026-06-10
💡 一句话要点
提出Damage-TriageFormer以解决灾后建筑损伤评估问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 建筑损伤评估 单幅图像分析 灾后响应 机器学习 深度学习
📋 核心要点
- 现有建筑损伤评估方法往往将损伤简化为单一的严重性等级,或依赖于缺乏的配对图像,限制了其在新兴灾害中的应用。
- 本文提出Damage-TriageFormer模型,基于单幅图像进行损伤类型评估,避免了对预事件图像的依赖,增强了评估的实用性。
- 实验结果显示,模型在无损建筑和完全结构崩溃的分类上分别达到了0.91和0.84的F1值,验证了其在实际应用中的有效性。
📝 摘要(中文)
建筑损伤评估对于灾后资源优先分配和恢复至关重要,但现有自动化方法通常将损伤简化为单一严重性等级,或依赖于往返事件的配对图像,后者在新兴灾害中往往不可用。本文提出Damage-TriageFormer,这是一种基于单幅图像的后事件模型,能够生成损伤类型而非严重性等级。我们贡献了DamageTriage-Bench,这是一个基于NOAA应急响应图像的新基准,涵盖了五种损伤类型,并提出了Damage-TriageFormer模型,扩展了DINOv3 ViT-L骨干网络,采用简单特征金字塔以实现更高分辨率的实例池化,结合两阶段门控损伤头和辅助严重性回归目标。我们的模型在验证集上获得了0.624的宏F1值,在保留的分层测试集上为0.619,尤其在操作性分诊需求高的场景中表现突出。
🔬 方法详解
问题定义:本文旨在解决灾后建筑损伤评估中的现有方法不足,尤其是对损伤的单一严重性等级评估和对配对图像的依赖问题。
核心思路:Damage-TriageFormer通过单幅图像生成损伤类型,提供更细致的损伤评估,支持针对性的应急响应和资源分配。
技术框架:该模型基于DINOv3 ViT-L骨干网络,结合简单特征金字塔以实现高分辨率的实例池化,采用两阶段门控损伤头和辅助严重性回归目标,形成完整的评估流程。
关键创新:Damage-TriageFormer的核心创新在于其能够在没有预事件图像的情况下,基于单幅后事件图像进行损伤类型评估,突破了传统方法的局限。
关键设计:模型设计中采用了简单特征金字塔以提高分辨率,损伤头采用两阶段门控机制,损失函数中引入了辅助严重性回归目标,以提升模型的整体性能。
🖼️ 关键图片
📊 实验亮点
实验结果表明,Damage-TriageFormer在验证集上获得了0.624的宏F1值,在保留的分层测试集上为0.619,尤其在无损建筑和完全结构崩溃的分类上分别达到了0.91和0.84的F1值,显示出其在实际应用中的强大能力。
🎯 应用场景
该研究的潜在应用领域包括自然灾害后的建筑损伤评估、应急响应和资源分配等。通过提供更准确的损伤类型评估,能够帮助决策者在灾后恢复过程中做出更有效的资源配置,提升救援效率。未来,该模型可扩展至其他类型的灾害评估和城市基础设施监测。
📄 摘要(原文)
Decision-relevant building damage assessment is critical for prioritizing resources and recovery after a disaster, yet most automated methods either flatten damage into a single severity scale (no damage, minor, major, destroyed) or require paired pre- and post-event imagery that is often unavailable for emerging hazards. This paper presents Damage-TriageFormer, a single-image, post-event, footprint-conditioned model that produces a damage typology rather than a severity scale. We contribute: (1) DamageTriage-Bench, a new benchmark built from NOAA Emergency Response Imagery across Hurricane Michael (2018), Hurricane Helene (2024), and the 2025 Los Angeles wildfire complex, with five typology classes that distinguish roof damage from structural damage and, within each, partial from total extent; and (2) Damage-TriageFormer, which extends a DINOv3 ViT-L backbone with a Simple Feature Pyramid for higher-resolution instance pooling, a two-stage gated damage head, and an auxiliary severity-regression objective. Our model achieves macro F1 of 0.624 on validation and 0.619 on a held-out stratified test set, performing strongest where operational triage needs it most, with per-class F1 of 0.91 and 0.84 on undamaged buildings and total structural collapse, respectively. While the rare Total Roof Damage class remains difficult due to its limited examples and an inherently ambiguous label boundary, our results show that single-image post-event imagery can support actionable building damage typing, enabling targeted emergency response and resource allocation without a pre-event reference.