GraVoS: Voxel Selection for 3D Point-Cloud Detection

Oren Shrout, Yizhak Ben-Shabat, Ayellet Tal

CVPR 2023 pp. 21684-21693

doi:10.1109/CVPR52729.2023.02077 /cvpr/2023/shrout2023cvpr-gravos/

Abstract

3D object detection within large 3D scenes is challenging not only due to the sparse and irregular 3D point clouds, but also due to both the extreme foreground-background scene imbalance and class imbalance. A common approach is to add ground-truth objects from other scenes. Differently, we propose to modify the scenes by removing elements (voxels), rather than adding ones. Our approach selects the "meaningful" voxels, in a manner that addresses both types of dataset imbalance. The approach is general and can be applied to any voxel-based detector, yet the meaningfulness of a voxel is network-dependent. Our voxel selection is shown to improve the performance of several prominent 3D detection methods.

PDF CVPR Semantic Scholar

Cite

Text

Shrout et al. "GraVoS: Voxel Selection for 3D Point-Cloud Detection." Conference on Computer Vision and Pattern Recognition, 2023. doi:10.1109/CVPR52729.2023.02077

Markdown

[Shrout et al. "GraVoS: Voxel Selection for 3D Point-Cloud Detection." Conference on Computer Vision and Pattern Recognition, 2023.](https://mlanthology.org/cvpr/2023/shrout2023cvpr-gravos/) doi:10.1109/CVPR52729.2023.02077

BibTeX

@inproceedings{shrout2023cvpr-gravos,
  title     = {{GraVoS: Voxel Selection for 3D Point-Cloud Detection}},
  author    = {Shrout, Oren and Ben-Shabat, Yizhak and Tal, Ayellet},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2023},
  pages     = {21684-21693},
  doi       = {10.1109/CVPR52729.2023.02077},
  url       = {https://mlanthology.org/cvpr/2023/shrout2023cvpr-gravos/}
}