Day41 - Neck中C3/C2f后添加一层GAM在VisDrone上的表现
yolov5n/s/m和yolov8n/s仅在neck的C3/C2f模块后添加一层GAM,在VisDrone2019上训练200epochs Yolov5n-gam Parameters: 3.336M Class Precision Recall ...
彩虹岛
Day40 - Backbone中C3/C2f后添加一层CBAM在VisDrone上的表现
yolov5n/s/m和yolov8n/s仅在backbone的C3/C2f模块后添加一层CBAM,在VisDrone2019上训练200epochs Yolov5n-cbam Parameters: 2.742M Class Precision Recall ...
Day39 - Neck中C3/C2f后添加一层CBAM在VisDrone上的表现
yolov5n/s/m和yolov8n/s仅在neck的C3/C2f模块后添加一层CBAM,在VisDrone2019上训练200epochs Yolov5n-cbam Parameters: 2.758M Class Precision Recall ...
Day38 - Neck中C3/C2f后添加一层CBAM在FLIR上的表现
yolov5n/s/m和yolov8n/s仅在neck的C3/C2f模块后添加一层CBAM,在FLIR上训练200epochs Yolov5n-cbam Parameters: 2.758M Class Precision Recall mA...
Day37 - Backbone中C3/C2f后添加一层CBAM在FLIR上的表现
yolov5n/s/m和yolov8n/s仅在backbone的C3/C2f模块后添加一层CBAM,在FLIR上训练200epochs Yolov5n-cbam Parameters: 2.742M Class Precision Recall ...
Day36 - 删除P5检测头+添加P2检测头（P234）在VisDrone上的表现
yolov5n/s/m和yolov8n/s删除P5检测头+添加P2检测头,在VisDrone2019上训练200epochs Yolov5n-p234 Parameters: 1.956M Class Precision Recall mAP...
Day35 - 删除P5检测头+添加P2检测头（P234）在FLIR上的表现
yolov5n/s/m和yolov8n/s删除P5检测头+添加P2检测头,在FLIR上训练200epochs Yolov5n-p234 Parameters: 1.956M Class Precision Recall mAP50 ...
Day34 - NWD在FLIR上的表现
yolov5n/s/m和yolov8n/s仅使用NWD损失函数,在FLIR上训练200epochs Yolov5n-nwd Parameters: 2.655M Class Precision Recall mAP50 m...
Day33 - NWD在VisDrone上的表现
yolov5n/s/m和yolov8n/s仅使用NWD损失函数,在VisDrone2019上训练200epochs Yolov5n-nwd Parameters: 2.655M Class Precision Recall mAP50 ...
Day32 - P2改进在VisDrone上的表现
yolov5n/s/m和yolov8n/s仅添加p2检测头,在VisDrone上训练200epochs Yolov5n-p2 Parameters: 2.838M Class Precision Recall mAP50 ...