Boost 3-D Object Detection via Point Clouds Segmentation and Fused 3-D GIoU-L Loss

Yaran Chen, Haoran Li, Ruiyuan Gao, Dongbin Zhao*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

30 Citations (Scopus)

Abstract

The 3-D object detection is crucial for many real-world applications, attracting many researchers' attention. Beyond 2-D object detection, 3-D object detection usually needs to extract appearance, depth, position, and orientation information from light detection and ranging (LiDAR) and camera sensors. However, due to more degrees of freedom and vertices, existing detection methods that directly transform from 2-D to 3-D still face several challenges, such as exploding increase of anchors' number and inefficient or hard-to-optimize objective. To this end, we present a fast segmentation method for 3-D point clouds to reduce anchors, which can largely decrease the computing cost. Moreover, taking advantage of 3-D generalized Intersection of Union (GIoU) and L1 losses, we propose a fused loss to facilitate the optimization of 3-D object detection. A series of experiments show that the proposed method has alleviated the abovementioned issues effectively.

Original languageEnglish
Pages (from-to)762-773
Number of pages12
JournalIEEE Transactions on Neural Networks and Learning Systems
Volume33
Issue number2
DOIs
Publication statusPublished - 1 Feb 2022

Keywords

  • 3-D object detection
  • generalized Intersection of Union (GIoU) loss
  • segmentation

Fingerprint

Dive into the research topics of 'Boost 3-D Object Detection via Point Clouds Segmentation and Fused 3-D GIoU-L Loss'. Together they form a unique fingerprint.

Cite this