A Few-Shot Attention Recurrent Residual U-Net for Crack Segmentation
This work addresses the need for dynamic adaptation in automated visual inspection of road infrastructure, though it is incremental as it builds on existing U-Net architectures.
The paper tackles the problem of automated road crack segmentation by introducing a few-shot learning paradigm that dynamically adapts to user feedback, achieving state-of-the-art performance with improved Dice and IoU metrics on a new public dataset.
Recent studies indicate that deep learning plays a crucial role in the automated visual inspection of road infrastructures. However, current learning schemes are static, implying no dynamic adaptation to users' feedback. To address this drawback, we present a few-shot learning paradigm for the automated segmentation of road cracks, which is based on a U-Net architecture with recurrent residual and attention modules (R2AU-Net). The retraining strategy dynamically fine-tunes the weights of the U-Net as a few new rectified samples are being fed into the classifier. Extensive experiments show that the proposed few-shot R2AU-Net framework outperforms other state-of-the-art networks in terms of Dice and IoU metrics, on a new dataset, named CrackMap, which is made publicly available at https://github.com/ikatsamenis/CrackMap.