Pavement crack detection with hybrid-window attentive vision transformers
International Journal of Applied Earth Observation and Geoinformation(2023)
摘要
•A multiscale token embedding strategy for token representation generation.•A hybrid-window vision transformer architecture for feature semantic extraction.•A weighted multi-head self-attention scheme for feature encoding promotion.•A high-resolution vision transformer model for pavement crack detection.
更多查看译文
关键词
Crack detection,Pavement distress,Attentive vision transformer,Feature attention,High-resolution network,Deep learning
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要