TornadoNet: Real-Time Building Damage Detection with Ordinal Supervision

2026-03-12

AI Summary

This study addresses the challenge of real-time detection and accurate multi-level damage assessment of buildings in post-disaster street-view imagery. We establish the first real-time benchmark for street-level building damage evaluation, systematically evaluating the performance of YOLO-family CNNs and Transformer-based models such as RT-DETR on a five-class damage classification task aligned with the IN-CORE standard. We propose a novel soft ordinal classification objective combined with an explicit ordinal distance penalty, revealing for the first time the synergistic impact of model architecture and ordinal supervision on grading performance. Experiments show that YOLO achieves 46.05% mAP@0.5 at 276 FPS, while RT-DETR attains a 91.15% ordinal Top-1 accuracy, reduces MAOE to 0.56, and improves mAP by 4.8 percentage points, significantly enhancing both accuracy and ordinal consistency.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• 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

We present TornadoNet, a comprehensive benchmark for automated street-level building damage assessment evaluating how modern real-time object detection architectures and ordinal-aware supervision strategies perform under realistic post-disaster conditions. TornadoNet provides the first controlled benchmark demonstrating how architectural design and loss formulation jointly influence multi-level damage detection from street-view imagery, delivering methodological insights and deployable tools for disaster response. Using 3,333 high-resolution geotagged images and 8,890 annotated building instances from the 2021 Midwest tornado outbreak, we systematically compare CNN-based detectors from the YOLO family against transformer-based models (RT-DETR) for multi-level damage detection. Models are trained under standardized protocols using a five-level damage classification framework based on IN-CORE damage states, validated through expert cross-annotation. Baseline experiments reveal complementary architectural strengths. CNN-based YOLO models achieve highest detection accuracy and throughput, with larger variants reaching 46.05% mAP@0.5 at 66-276 FPS on A100 GPUs. Transformer-based RT-DETR models exhibit stronger ordinal consistency, achieving 88.13% Ordinal Top-1 Accuracy and MAOE of 0.65, indicating more reliable severity grading despite lower baseline mAP. To align supervision with the ordered nature of damage severity, we introduce soft ordinal classification targets and evaluate explicit ordinal-distance penalties. RT-DETR trained with calibrated ordinal supervision achieves 44.70% mAP@0.5, a 4.8 percentage-point improvement, with gains in ordinal metrics (91.15% Ordinal Top-1 Accuracy, MAOE = 0.56). These findings establish that ordinal-aware supervision improves damage severity estimation when aligned with detector architecture. Model & Data: https://github.com/crumeike/TornadoNet

Problem

• building damage detection
• ordinal supervision
• real-time assessment
• disaster response
• street-level imagery

Innovation

• ordinal supervision
• real-time damage detection
• TornadoNet
• RT-DETR
• building damage assessment