WP-CrackNet: A Collaborative Adversarial Learning Framework for End-to-End Weakly-Supervised Road Crack Detection

2025-10-20

AI Summary

To address the high cost of pixel-level annotations in road crack detection, this paper proposes an end-to-end weakly supervised method that achieves high-precision pixel-level localization using only image-level labels. The method introduces a collaborative adversarial learning framework, incorporating a novel path-aware attention module to better model crack structures and a center-enhanced CAM consistency module that enforces Gaussian-weighted constraints to improve the completeness and localization robustness of Class Activation Maps (CAMs). Furthermore, it integrates feature inferability reconstruction with spatial-channel joint attention to enable multi-scale semantic fusion and pseudo-label refinement. Evaluated on a newly constructed road dataset, the proposed approach matches the performance of fully supervised baselines and significantly outperforms existing weakly supervised methods. This work provides an efficient and scalable technical solution for large-scale intelligent infrastructure inspection.

Technology Category

• Computer Vision: Segmentation
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Intelligent Robots: Localization, Mapping, and Navigation

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Road crack detection is essential for intelligent infrastructure maintenance in smart cities. To reduce reliance on costly pixel-level annotations, we propose WP-CrackNet, an end-to-end weakly-supervised method that trains with only image-level labels for pixel-wise crack detection. WP-CrackNet integrates three components: a classifier generating class activation maps (CAMs), a reconstructor measuring feature inferability, and a detector producing pixel-wise road crack detection results. During training, the classifier and reconstructor alternate in adversarial learning to encourage crack CAMs to cover complete crack regions, while the detector learns from pseudo labels derived from post-processed crack CAMs. This mutual feedback among the three components improves learning stability and detection accuracy. To further boost detection performance, we design a path-aware attention module (PAAM) that fuses high-level semantics from the classifier with low-level structural cues from the reconstructor by modeling spatial and channel-wise dependencies. Additionally, a center-enhanced CAM consistency module (CECCM) is proposed to refine crack CAMs using center Gaussian weighting and consistency constraints, enabling better pseudo-label generation. We create three image-level datasets and extensive experiments show that WP-CrackNet achieves comparable results to supervised methods and outperforms existing weakly-supervised methods, significantly advancing scalable road inspection. The source code package and datasets are available at https://mias.group/WP-CrackNet/.

Problem

• Reduces reliance on costly pixel-level crack annotations
• Improves weakly-supervised road crack detection accuracy
• Generates better pseudo-labels through adversarial learning modules

Innovation

• Adversarial learning integrates classifier and reconstructor components
• Path-aware attention fuses high-level semantics with low-level cues
• Center-enhanced consistency refines crack maps using Gaussian weighting