DAP-MAE: Domain-Adaptive Point Cloud Masked Autoencoder for Effective Cross-Domain Learning

2025-10-24

AI Summary

To address the weak knowledge transfer capability of Masked Autoencoders (MAEs) under scarce cross-domain point cloud data and significant domain discrepancies, this paper proposes Domain-Adaptive MAE (DAP-MAE). Methodologically, it introduces domain adaptation into the point cloud MAE framework for the first time, designing a dual-mode heterogeneous domain adapter and a domain feature generator to enable adaptive cross-domain knowledge fusion during pre-training and task-oriented feature enhancement during fine-tuning. The approach integrates cross-domain contrastive learning with dynamic feature fusion, supporting single pre-training for multiple downstream tasks. Evaluated on ScanObjectNN and Bosphorus, DAP-MAE achieves 95.18% and 88.45% classification accuracy, respectively—demonstrating substantial improvements in cross-domain generalization and multi-task adaptability.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Compared to 2D data, the scale of point cloud data in different domains available for training, is quite limited. Researchers have been trying to combine these data of different domains for masked autoencoder (MAE) pre-training to leverage such a data scarcity issue. However, the prior knowledge learned from mixed domains may not align well with the downstream 3D point cloud analysis tasks, leading to degraded performance. To address such an issue, we propose the Domain-Adaptive Point Cloud Masked Autoencoder (DAP-MAE), an MAE pre-training method, to adaptively integrate the knowledge of cross-domain datasets for general point cloud analysis. In DAP-MAE, we design a heterogeneous domain adapter that utilizes an adaptation mode during pre-training, enabling the model to comprehensively learn information from point clouds across different domains, while employing a fusion mode in the fine-tuning to enhance point cloud features. Meanwhile, DAP-MAE incorporates a domain feature generator to guide the adaptation of point cloud features to various downstream tasks. With only one pre-training, DAP-MAE achieves excellent performance across four different point cloud analysis tasks, reaching 95.18% in object classification on ScanObjectNN and 88.45% in facial expression recognition on Bosphorus.

Problem

• Limited point cloud data across domains for training
• Mixed domain knowledge misalignment with downstream tasks
• Adaptively integrating cross-domain datasets for point cloud analysis

Innovation

• Domain-adaptive masked autoencoder for cross-domain point clouds
• Heterogeneous domain adapter with dual training modes
• Domain feature generator enhances downstream task adaptation