MANet: End-to-End Learning for Point Cloud Based on Robust Pointpillar and Multiattention

Table 3

Results on the KITTI test 3D detection benchmark.

Methods	Mod	Bev mAP	Car			Pedestrian			Cyclist
			Easy	Mod	Hard	Easy	Mod	Hard	Easy	Mod	Hard
MV3D [38]	Lidar&img	N/A	71.09	62.35	55.12
F-P [9]		57.35	81.20	70.39	62.19	51.21	44.89	40.23	71.96	56.77	50.39
VoxelNet [3]	Lidar	49.05	77.47	65.11	57.73	39.48	33.69	31.5	61.22	48.36	44.37
SECOND [39]		56.69	83.13	73.66	66.20	51.07	42.56	37.29	70.51	53.85	46.90
Pp [6]		59.20	79.05	74.99	68.30	52.08	43.53	41.49	75.78	59.07	52.92
TANet [7]		62	83.81	75.38	67.66	54.92	46.67	38.63	73.93	59.60	53.59
MANet		65.11	83.47	78.85	71.89	56.02	51.74	45.58	80.08	64.74	60.79