EURASIP Journal on Advances in Signal Processing

Table 2 The mAP (%) comparison of the proposed method with YOLOv3 on KITTI

From: Multiclass objects detection algorithm using DarkNet-53 and DenseNet for intelligent vehicles

Method	Backbone	Input size	AP (%)					mAP (%)
Method	Backbone	Input size	Car	Pedestrian	Cyclist	Van	Truck	mAP (%)
YOLOv3	DarkNet53	608 × 608	92.22	78.97	83.24	92.99	94.30	88.34
Ours	DarkNet53	608 × 608	92.74	81.96	89.83	94.50	96.61	91.13

Back to article page