MaskLRF: Self-Supervised Pretraining via Masked Autoencoding of Local Reference Frames for Rotation-Invariant 3D Point Set Analysis

Following the successes in the fields of vision and language, self-supervised pretraining via masked autoencoding of 3D point set data, or Masked Point Modeling (MPM), has achieved state-of-the-art accuracy in various downstream tasks. However, current MPM methods lack a property essential for 3D po...

Полное описание

Библиографические подробности

Главный автор:	Takahiko Furuya
Формат:	Artigo
Язык:	Inglês
Опубликовано:	IEEE 2024-01-01
Серии:	IEEE Access
Предметы:	3D point cloud deep learning masked autoencoding representation learning self-supervised pretraining
Online-ссылка:	https://ieeexplore.ieee.org/document/10536090/
Метки:	Добавить метку Нет меток, Требуется 1-ая метка записи!