Curriculumformer: Taming Curriculum Pre-Training for Enhanced 3-D Point Cloud Understanding

Learning universal representations of 3-D point clouds is essential for reducing the need for manual annotation of large-scale and irregular point cloud datasets. The current modus operandi for representative learning is self-supervised learning, which has shown great potential for improving point c...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE transaction on neural networks and learning systems 2024-06, Vol.PP, p.1-15
Hauptverfasser:	Fei, Ben, Luo, Tianyue, Yang, Weidong, Liu, Liwen, Zhang, Rui, He, Ying
Format:	Artikel
Sprache:	eng
Schlagworte:	3-D representation learning Accuracy curriculum learning Data models Geometry Point cloud compression point clouds Representation learning self-supervised learning Task analysis transformer Transformers
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!