Options
Toward a robust sensor fusion step for 3D object detection on corrupted data
Publikationstyp
Journal Article
Date Issued
2023-11-01
Sprache
English
Author(s)
Volume
8
Issue
11
Start Page
7018
End Page
7025
Citation
IEEE Robotics and Automation Letters 8 (11): 7018-7025 (2023-11-01)
Publisher DOI
Scopus ID
Publisher
IEEE
Multimodal sensor fusion methods for 3D object detection have been revolutionizing the autonomous driving research field. Nevertheless, most of these methods heavily rely on dense LiDAR data and accurately calibrated sensors which is often not the case in real-world scenarios. Data from LiDAR and cameras often come misaligned due to the miscalibration, decalibration, or different frequencies of the sensors. Additionally, some parts of the LiDAR data may be occluded and parts of the data may be missing due to hardware malfunction or weather conditions. This work presents a novel fusion step that addresses data corruptions and makes sensor fusion for 3D object detection more robust. Through extensive experiments, we demonstrate that our method performs on par with state-of-the-art approaches on normal data and outperforms them on misaligned data.
Subjects
deep learning for visual perception
Object detection
segmentation and categorization
sensor fusion
DDC Class
670: Manufacturing
600: Technology