Joint data and feature augmentation for self-supervised representation learning on point clouds

Joint data and feature augmentation for self-supervised representation learning on point clouds

To deal with the exhausting annotations, self-supervised representation learning from unlabeled point clouds has drawn much attention, especially centered on augmentation-based contrastive methods. However, specific augmentations hardly produce sufficient transferability to high-level tasks on diffe...

Cijeli opis

Bibliografski detalji
Glavni autori: Zhuheng Lu, Yuewei Dai, Weiqing Li, Zhiyong Su
Format: Članak
Jezik:English
Izdano: Elsevier 2023-10-01
Serija:Graphical Models
Teme:
Contrastive learning
Self-supervised learning
Point cloud
Representation learning
Online pristup:http://www.sciencedirect.com/science/article/pii/S1524070323000188

Internet

http://www.sciencedirect.com/science/article/pii/S1524070323000188

Slični predmeti