Point Cloud Mixture-of-Domain-Experts Model for 3D Self-supervised Learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zha, Yaohua, Dai, Tao, Guo, Hang, Wang, Yanzi, Chen, Bin, Chen, Ke, Xia, Shu-Tao
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915319013965824
author Zha, Yaohua
Dai, Tao
Guo, Hang
Wang, Yanzi
Chen, Bin
Chen, Ke
Xia, Shu-Tao
author_facet Zha, Yaohua
Dai, Tao
Guo, Hang
Wang, Yanzi
Chen, Bin
Chen, Ke
Xia, Shu-Tao
contents Point clouds, as a primary representation of 3D data, can be categorized into scene domain point clouds and object domain point clouds. Point cloud self-supervised learning (SSL) has become a mainstream paradigm for learning 3D representations. However, existing point cloud SSL primarily focuses on learning domain-specific 3D representations within a single domain, neglecting the complementary nature of cross-domain knowledge, which limits the learning of 3D representations. In this paper, we propose to learn a comprehensive Point cloud Mixture-of-Domain-Experts model (Point-MoDE) via a block-to-scene pre-training strategy. Specifically, we first propose a mixture-of-domain-expert model consisting of scene domain experts and multiple shared object domain experts. Furthermore, we propose a block-to-scene pretraining strategy, which leverages the features of point blocks in the object domain to regress their initial positions in the scene domain through object-level block mask reconstruction and scene-level block position regression. By integrating the complementary knowledge between object and scene, this strategy simultaneously facilitates the learning of both object-domain and scene-domain representations, leading to a more comprehensive 3D representation. Extensive experiments in downstream tasks demonstrate the superiority of our model.
format Preprint
id arxiv_https___arxiv_org_abs_2410_09886
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Point Cloud Mixture-of-Domain-Experts Model for 3D Self-supervised Learning
Zha, Yaohua
Dai, Tao
Guo, Hang
Wang, Yanzi
Chen, Bin
Chen, Ke
Xia, Shu-Tao
Computer Vision and Pattern Recognition
Point clouds, as a primary representation of 3D data, can be categorized into scene domain point clouds and object domain point clouds. Point cloud self-supervised learning (SSL) has become a mainstream paradigm for learning 3D representations. However, existing point cloud SSL primarily focuses on learning domain-specific 3D representations within a single domain, neglecting the complementary nature of cross-domain knowledge, which limits the learning of 3D representations. In this paper, we propose to learn a comprehensive Point cloud Mixture-of-Domain-Experts model (Point-MoDE) via a block-to-scene pre-training strategy. Specifically, we first propose a mixture-of-domain-expert model consisting of scene domain experts and multiple shared object domain experts. Furthermore, we propose a block-to-scene pretraining strategy, which leverages the features of point blocks in the object domain to regress their initial positions in the scene domain through object-level block mask reconstruction and scene-level block position regression. By integrating the complementary knowledge between object and scene, this strategy simultaneously facilitates the learning of both object-domain and scene-domain representations, leading to a more comprehensive 3D representation. Extensive experiments in downstream tasks demonstrate the superiority of our model.
title Point Cloud Mixture-of-Domain-Experts Model for 3D Self-supervised Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.09886