Earf-YOLO: An Efficient Attention Receptive Field Model for Recognizing Symbols of Zhuang Minority Patterns

Table 5

In VOC dataset, the comparative results of YOLOv3, YOLOv4, and YOLOv5 with the main contributions of the proposed model are shown.

Methods	AP(%)
YOLOv3 416	79.3
YOLOv4 640	88.9
YOLOv5s	86.8
YOLOv3+GLocalT + SRFB	82.5
YOLOv4+GLocalT + SRFB	90.1
YOLOv5+GLocalT + SRFB	88.4