Benutzer: Gast  Login
Mehr Felder
Einfache Suche
Titel:

3D-VField: Adversarial Augmentation of Point Clouds for Domain Generalization in 3D Object Detection

Dokumenttyp:
Konferenzbeitrag
Autor(en):
Alexander Lehner; Stefano Gasperini; Alvaro Marcos-Ramiro; Michael Schmidt; Mohammad-Ali Nikouei Mahani; Nassir Navab; Benjamin Busam; Federico Tombari
Abstract:
As 3D object detection on point clouds relies on the geometrical relationships between the points, non-standard object shapes can hinder a method’s detection capability. However, in safety-critical settings, robustness to out-of-domain and long-tail samples is fundamental to circumvent dangerous issues, such as the misdetection of damaged or rare cars. In this work, we substantially improve the generalization of 3D object detectors to out-of-domain data by deforming point clouds during training....     »
Stichworte:
adversarial augmentation; domain generalization; out-of-distribution; point clouds; 3D object detection
Dewey-Dezimalklassifikation:
000 Informatik, Wissen, Systeme
Kongress- / Buchtitel:
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Jahr:
2022
Monat:
Jun
Reviewed:
ja
Volltext / DOI:
doi:10.1109/CVPR52688.2022.01678
WWW:
Paper via CVPR 2022 Open Access
Hinweise:
The first two authors contributed equally.
Copyright Informationen:
Copyright with IEEE.
 BibTeX