dengue-detection

This paper proposes a YOLO- and CLIP-based vision-language framework to classify mosquito flight frames for Dengue virus detection, achieving 98.54% accuracy and 99.91% sensitivity at frame level, with complete video-level performance after temporal aggregation.