Saved in:
Bibliographic Details
Main Authors: Xu, Shaocong, Li, Pengfei, Sun, Qianpu, Liu, Xinyu, Li, Yang, Guo, Shihui, Wang, Zhen, Jiang, Bo, Wang, Rui, Sheng, Kehua, Zhang, Bo, Jiang, Li, Zhao, Hao, Chen, Yilun
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2309.10230
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916531401654272
author Xu, Shaocong
Li, Pengfei
Sun, Qianpu
Liu, Xinyu
Li, Yang
Guo, Shihui
Wang, Zhen
Jiang, Bo
Wang, Rui
Sheng, Kehua
Zhang, Bo
Jiang, Li
Zhao, Hao
Chen, Yilun
author_facet Xu, Shaocong
Li, Pengfei
Sun, Qianpu
Liu, Xinyu
Li, Yang
Guo, Shihui
Wang, Zhen
Jiang, Bo
Wang, Rui
Sheng, Kehua
Zhang, Bo
Jiang, Li
Zhao, Hao
Chen, Yilun
contents LiDAR-based semantic scene understanding is an important module in the modern autonomous driving perception stack. However, identifying outlier points in a LiDAR point cloud is challenging as LiDAR point clouds lack semantically-rich information. While former SOTA methods adopt heuristic architectures, we revisit this problem from the perspective of Selective Classification, which introduces a selective function into the standard closed-set classification setup. Our solution is built upon the basic idea of abstaining from choosing any inlier categories but learns a point-wise abstaining penalty with a margin-based loss. Apart from learning paradigms, synthesizing outliers to approximate unlimited real outliers is also critical, so we propose a strong synthesis pipeline that generates outliers originated from various factors: object categories, sampling patterns and sizes. We demonstrate that learning different abstaining penalties, apart from point-wise penalty, for different types of (synthesized) outliers can further improve the performance. We benchmark our method on SemanticKITTI and nuScenes and achieve SOTA results. Codes are available at https://github.com/Daniellli/LiON/.
format Preprint
id arxiv_https___arxiv_org_abs_2309_10230
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle LiON: Learning Point-wise Abstaining Penalty for LiDAR Outlier DetectioN Using Diverse Synthetic Data
Xu, Shaocong
Li, Pengfei
Sun, Qianpu
Liu, Xinyu
Li, Yang
Guo, Shihui
Wang, Zhen
Jiang, Bo
Wang, Rui
Sheng, Kehua
Zhang, Bo
Jiang, Li
Zhao, Hao
Chen, Yilun
Computer Vision and Pattern Recognition
LiDAR-based semantic scene understanding is an important module in the modern autonomous driving perception stack. However, identifying outlier points in a LiDAR point cloud is challenging as LiDAR point clouds lack semantically-rich information. While former SOTA methods adopt heuristic architectures, we revisit this problem from the perspective of Selective Classification, which introduces a selective function into the standard closed-set classification setup. Our solution is built upon the basic idea of abstaining from choosing any inlier categories but learns a point-wise abstaining penalty with a margin-based loss. Apart from learning paradigms, synthesizing outliers to approximate unlimited real outliers is also critical, so we propose a strong synthesis pipeline that generates outliers originated from various factors: object categories, sampling patterns and sizes. We demonstrate that learning different abstaining penalties, apart from point-wise penalty, for different types of (synthesized) outliers can further improve the performance. We benchmark our method on SemanticKITTI and nuScenes and achieve SOTA results. Codes are available at https://github.com/Daniellli/LiON/.
title LiON: Learning Point-wise Abstaining Penalty for LiDAR Outlier DetectioN Using Diverse Synthetic Data
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2309.10230