RDD4D: 4D Attention-Guided Road Damage Detection And Classification

2025-01-06

AI Summary

To address performance limitations in multi-scale, multi-type road damage detection caused by data scarcity, this paper introduces DRDD—the first benchmark dataset featuring rich morphological and scale variations—and proposes RDD4D, a novel detection model. Methodologically, RDD4D integrates a Transformer-enhanced CNN backbone with a multi-scale feature fusion architecture. Its core innovation is the Attention4D module, the first to jointly incorporate four-dimensional attention (spatial, channel, scale, and semantic) alongside positional encoding and the Talking-Head mechanism for cross-scale feature refinement. Evaluated on DRDD, RDD4D achieves an mAP of 0.445 (with 0.458 AP for large cracks), substantially outperforming state-of-the-art methods; it further improves mAP by approximately 0.21 on CrackTinyNet. The code, pre-trained models, and the DRDD dataset are fully open-sourced.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Road damage detection and assessment are crucial components of infrastructure maintenance. However, current methods often struggle with detecting multiple types of road damage in a single image, particularly at varying scales. This is due to the lack of road datasets with various damage types having varying scales. To overcome this deficiency, first, we present a novel dataset called Diverse Road Damage Dataset (DRDD) for road damage detection that captures the diverse road damage types in individual images, addressing a crucial gap in existing datasets. Then, we provide our model, RDD4D, that exploits Attention4D blocks, enabling better feature refinement across multiple scales. The Attention4D module processes feature maps through an attention mechanism combining positional encoding and"Talking Head"components to capture local and global contextual information. In our comprehensive experimental analysis comparing various state-of-the-art models on our proposed, our enhanced model demonstrated superior performance in detecting large-sized road cracks with an Average Precision (AP) of 0.458 and maintained competitive performance with an overall AP of 0.445. Moreover, we also provide results on the CrackTinyNet dataset; our model achieved around a 0.21 increase in performance. The code, model weights, dataset, and our results are available on href{https://github.com/msaqib17/Road_Damage_Detection}{https://github.com/msaqib17/Road_Damage_Detection}.

Problem

• Road Damage Detection
• Diverse Dataset
• Model Training

Innovation

• RDD4D model
• Attention4D mechanism
• DRDD dataset