PaveSync: A Unified and Comprehensive Dataset for Pavement Distress Analysis and Classification

2025-12-22

AI Summary

Current pavement defect detection suffers from poor cross-scenario generalization due to the absence of standardized, globally representative datasets. To address this, we introduce PavementBench—the first standardized benchmark dataset covering seven countries, comprising 52,747 high-resolution images and 135,277 bounding-box annotations. It rigorously defines 13 distinct pavement distress categories and standardizes annotation formats and evaluation protocols. We further propose a novel cross-source consistency framework for defect definitions, enabling zero-shot cross-environment transfer. Extensive benchmarking is conducted across state-of-the-art detectors—including YOLOv8–v12, Faster R-CNN, and DETR—demonstrating substantial improvements in robustness and generalization across diverse real-world scenarios. PavementBench fills a critical gap in reproducible, comparable AI training benchmarks for pavement analysis, establishing a rigorous foundation for fair model evaluation and advancement of intelligent infrastructure inspection.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Evaluation and Analysis
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Automated pavement defect detection often struggles to generalize across diverse real-world conditions due to the lack of standardized datasets. Existing datasets differ in annotation styles, distress type definitions, and formats, limiting their integration for unified training. To address this gap, we introduce a comprehensive benchmark dataset that consolidates multiple publicly available sources into a standardized collection of 52747 images from seven countries, with 135277 bounding box annotations covering 13 distinct distress types. The dataset captures broad real-world variation in image quality, resolution, viewing angles, and weather conditions, offering a unique resource for consistent training and evaluation. Its effectiveness was demonstrated through benchmarking with state-of-the-art object detection models including YOLOv8-YOLOv12, Faster R-CNN, and DETR, which achieved competitive performance across diverse scenarios. By standardizing class definitions and annotation formats, this dataset provides the first globally representative benchmark for pavement defect detection and enables fair comparison of models, including zero-shot transfer to new environments.

Problem

• Standardizes diverse pavement distress datasets for unified analysis
• Enables consistent training and evaluation across varied real-world conditions
• Provides a global benchmark for fair model comparison in defect detection

Innovation

• Standardized dataset consolidates multiple sources globally
• Benchmarked with state-of-the-art object detection models
• Enables fair comparison and zero-shot transfer to new environments