SPDNet: A Real-Time Passenger Detection Method Based on Attention Mechanism in Subway Station Scenes

Table 2

Comparison of different attention methods on KITTI from the perspective of mean Average Precision (mAP), Person Average Precision (Person AP), Supplemental Module Parameters (Module Param), and Floating-Point Operations per second (FLOPs).

Network model	[email protected]	Person ([email protected])	Module Param	FLOPs
YOLOv4 (baseline)	86.3%	78.3%	0	128.46B
YOLOv4-ECA	87.7%	76.3%	5
YOLOv4-SE	88.4%	76.8%	32768
SPDNet	89.1%	79.7%	32771