3 results
Now showing1 - 3 of 3
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection With Sparse LiDAR and Large Domain GapsIn this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains.Publicationtype: Journal Article23 43 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Toward a robust sensor fusion step for 3D object detection on corrupted dataMultimodal 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.Publicationtype: Journal Article23 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Applying 3D Object Detection from Self-Driving Cars to Mobile Robots: A Survey and Experiments(2023-04); ; 3D object detection is crucial for the safety and reliability of mobile robots. Mobile robots must understand dynamic environments to operate safely and successfully carry out their tasks. However, most of the open-source datasets and methods are built for autonomous driving. In this paper, we present a detailed review of available 3D object detection methods, focusing on the ones that could be easily adapted and used on mobile robots. We show that the methods do not perform well if used off-the-shelf on mobile robots or are too computationally expensive to run on mobile robotic platforms. Therefore, we propose a domain adaptation approach to use publicly available data to retrain the perception modules of mobile robots, resulting in higher performance. Finally, we run the tests on the real-world robot and provide data for testing our approach.Publicationtype: Conference Paper28