Multi-view Structural Convolution Network for Domain-Invariant Point Cloud Recognition of Autonomous Vehicles

TL;DR

This paper introduces MSCN, a novel neural network architecture for point cloud recognition in autonomous vehicles that maintains high accuracy across different domains by learning domain-invariant features.

Contribution

The paper proposes MSCN with Structural Convolution and Aggregation Layers, and a training strategy with unseen domain data to improve domain invariance in point cloud recognition.

Findings

Achieves 82.0% accuracy across domains

Surpasses PointTransformer by 15.8% in accuracy

Demonstrates robustness under real-world domain shifts

Abstract

Point cloud representation has recently become a research hotspot in the field of computer vision and has been utilized for autonomous vehicles. However, adapting deep learning networks for point cloud data recognition is challenging due to the variability in datasets and sensor technologies. This variability underscores the necessity for adaptive techniques to maintain accuracy under different conditions. In this paper, we present the Multi-View Structural Convolution Network (MSCN) designed for domain-invariant point cloud recognition. MSCN comprises Structural Convolution Layers (SCL) that extract local context geometric features from point clouds and Structural Aggregation Layers (SAL) that extract and aggregate both local and overall context features from point clouds. Furthermore, MSCN enhances feature robustness by training with unseen domain point clouds generated from the source domain, enabling the model to acquire domain-invariant representations. Extensive cross-domain experiments demonstrate that MSCN achieves an average accuracy of 82.0%, surpassing the strong baseline PointTransformer by 15.8%, confirming its effectiveness under real-world domain shifts. Our code is available at https://github.com/MLMLab/MSCN.