Geometry and Uncertainty-Aware 3D Point Cloud Class-Incremental Semantic Segmentation

Despite the significant recent progress made on 3D point cloud semantic segmentation, the current methods require training data for all classes at once, and are not suitable for real-life scenarios where new categories are being continuously discovered. Substantial memory storage and expensive re-tr...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) S. 21759 - 21768
Hauptverfasser:	Yang, Yuwei, Hayat, Munawar, Jin, Zhao, Ren, Chao, Lei, Yinjie
Format:	Tagungsbericht
Sprache:	Englisch
Veröffentlicht:	IEEE 01.06.2023
Schlagworte:	continual low-shot meta or long-tail learning Point cloud compression Semantic segmentation Semantics Solid modeling Three-dimensional displays Training Transfer Upper bound
ISSN:	1063-6919
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!