---
title: "Point Cloud Mixture-of-Domain-Experts Model for 3D Self-supervised Learning"
authors:
  - "Yaohua Zha"
  - "Tao Dai"
  - "Hang Guo"
  - "Yanzi Wang"
  - "Bin Chen"
  - "Ke Chen"
  - "Shu-Tao Xia"
doi: "10.24963/ijcai.2025/260"
url: "https://ijcai.org/proceedings/2025/260"
pdf: "https://ijcai.org/proceedings/2025/0260.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "2332-2340"
track: "Main Track"
keywords:
  - "Computer Vision: CV: Recognition (object detection, categorization)"
  - "Computer Vision: CV: 3D computer vision"
---

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

Yaohua Zha, Tao Dai, Hang Guo, Yanzi Wang, Bin Chen, Ke Chen, Shu-Tao Xia

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 2332-2340. DOI: [10.24963/ijcai.2025/260](https://doi.org/10.24963/ijcai.2025/260)

## Abstract

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.