Paper: https://doi.org/10.1007/s00530-026-02521-5

Post time Half hour(s) ago

Paper: https://doi.org/10.1007/s00530-026-02521-5

Authors: Z Liang, J Yutong, C Dengfeng, L Shipeng

Abstract: Crack detection is crucial for infrastructure maintenance, yet it poses significant visual challenges due to the subtle and irregular nature of cracks. Traditional deep learning models, while effective, are often too complex for deployment on resource-constrained devices. This paper introduces a Memory-Aware Dynamic Distillation (MDD) framework to address this gap. By leveraging an unsupervised difficulty assessment strategy, a Gradient-Aware Memory Bank, and adaptive review scheduling, MDD effectively transfers knowledge from a large teacher model to a lightweight student model. Experimental results on the Crack Segmentation dataset demonstrate that MDD achieves a 5% improvement in mAP50 over standard knowledge distillation, with a 95.1% reduction in parameters and a 1.9-fold increase in inference speed. Here, we show that MDD significantly mitigates catastrophic forgetting, enhancing the applicability of lightweight detection models in real-world scenarios. The code and dataset are available at https://github.com/Jiaoyut/MDD-Crack-Detection.git and https://doi.org/10.5281/zenodo.18059674.

Post time Half hour(s) ago

: This post has been completed
Completed attachments will be deleted within 24 hours.

[Other] Paper: https://doi.org/10.1007/s00530-026-02521-5

Best Answer

Same Category

Latest Reply

Announcement

Daily Top Contributors

[Other] Paper: https://doi.org/10.1007/s00530-026-02521-5 Copy

Best Answer

Same Category

Latest Reply

Announcement

Daily Top Contributors

[Other] Paper: https://doi.org/10.1007/s00530-026-02521-5