Chinese Journal of Mechanical Engineering

Table 3 Performance comparison of 2D object detection

From: Deep Learning Based Data Fusion for Sensor Fault Diagnosis and Tolerance in Autonomous Vehicles

Network	Running time (s)	Easy (%)	Moderate (%)	Hard (%)
Faster-RCNN	2	88.97	83.16	72.62
RetinaNet [28]	0.2	93.97	82.73	68.37
RefineNet [29]	0.2	91.91	81.01	65.67
MonoFENet [30]	0.15	91.68	84.63	76.71
ResNet-RRC	0.06	91.45	85.33	74.24
Fast-SSD	0.06	85.19	66.79	57.89
YOLO800	0.13	78.93	74.31	63.83
The new method	0.02	95.36	88.20	86.44

Back to article page