Skip to main content
Journal of Big Data

Table 3 3D detection on the 16-channel LiDAR mobile robot dataset

From: Out-of-distribution- and location-aware PointNets for real-time 3D road user detection without a GPU

Method

Cars AP (0.5)

Pedestrians AP (0.25)

Speed FPS

GPU

CPU

VoxelNet [3]

18.21

17.13

2.3

0.1

SECOND [4]

18.56

18.01

4.8

1.3

IA-SSD [15]

34.24

22.45

20.2

4.4

PointPillars [9]

19.22

18.88

18.1

2.3

Ours

34.25

15.69

77.2

16.1

  1. Bold font indicates the best FPS
  2. All methods are trained only with the KITTI dataset. GPU: Nvidia GTX 1060, CPU: 4.0 GHz Intel i5-7600K 5th generation
Back to article page