Modeling Multi-Granularity Context Information Flow for Pavement Crack Detection
Junbiao Pang, Baocheng Xiong, Jiaqi Wu
TL;DR
This work tackles pavement crack detection under challenging conditions (low contrast, irregular crack structure, weak continuity) by modeling multi-granularity context information flow. MGCrackNet combines a parallel dilated-convolution backbone with a Context Guidance module and MIL-based label alignment to fuse local and semantic cues across stages, supervised by a multi-stage loss. The authors release the large Bitumen Pavement Crack (BPC) dataset and demonstrate state-of-the-art performance on CFD, CrackTree, and BPC, achieving high $AP$ and favorable $F1$ on patch-level crack detection. The approach promises robust, scalable crack detection in noisy real-world road imagery and offers a practical path for patch-level annotation reduction.
Abstract
Crack detection has become an indispensable, interesting yet challenging task in the computer vision community. Specially, pavement cracks have a highly complex spatial structure, a low contrasting background and a weak spatial continuity, posing a significant challenge to an effective crack detection method. In this paper, we address these problems from a view that utilizes contexts of the cracks and propose an end-to-end deep learning method to model the context information flow. To precisely localize crack from an image, it is critical to effectively extract and aggregate multi-granularity context, including the fine-grained local context around the cracks (in spatial-level) and the coarse-grained semantics (in segment-level). Concretely, in Convolutional Neural Network (CNN), low-level features extracted by the shallow layers represent the local information, while the deep layers extract the semantic features. Additionally, a second main insight in this work is that the semantic context should be an guidance to local context feature. By the above insights, the proposed method we first apply the dilated convolution as the backbone feature extractor to model local context, then we build a context guidance module to leverage semantic context to guide local feature extraction at multiple stages. To handle label alignment between stages, we apply the Multiple Instance Learning (MIL) strategy to align the high-level feature to the low-level ones in the stage-wise context flow. In addition, compared with these public crack datasets, to our best knowledge, we release the largest, most complex and most challenging Bitumen Pavement Crack (BPC) dataset. The experimental results on the three crack datasets demonstrate that the proposed method performs well and outperforms the current state-of-the-art methods.