SECOND: Sparsely Embedded Convolutional Detection

Explore this paper's citation graph

Summary

An improved sparse convolution method for Voxel-based 3D convolutional networks is investigated, which significantly increases the speed of both training and inference and introduces a new form of angle loss regression to improve the orientation estimation performance.

Type
article
Published
2018-10-01
Cited by
3,657
References
36
Access
Open access

Keywords

Point cloud, Lidar, Computer science, Artificial intelligence, Inference

References

Cited by

Related papers