Multilateral Cascading Network for Semantic Segmentation of Large-Scale Outdoor Point Clouds

TL;DR

This paper introduces MCNet, a novel network architecture with attention and feature fusion modules, significantly improving semantic segmentation accuracy on large-scale outdoor point cloud datasets.

Contribution

The study proposes MCNet with MCAE and P-CSP modules, advancing feature learning and fusion for outdoor point cloud segmentation, outperforming existing methods.

Findings

Achieved 2.1% higher mIoU on SensatUrban dataset.

Improved small-sample object classification by 15.9%.

Outperformed state-of-the-art methods on benchmark datasets.

Abstract

Semantic segmentation of large-scale outdoor point clouds is of significant importance in environment perception and scene understanding. However, this task continues to present a significant research challenge, due to the inherent complexity of outdoor objects and their diverse distributions in real-world environments. In this study, we propose the Multilateral Cascading Network (MCNet) designed to address this challenge. The model comprises two key components: a Multilateral Cascading Attention Enhancement (MCAE) module, which facilitates the learning of complex local features through multilateral cascading operations; and a Point Cross Stage Partial (P-CSP) module, which fuses global and local features, thereby optimizing the integration of valuable feature information across multiple scales. Our proposed method demonstrates superior performance relative to state-of-the-art approaches across two widely recognized benchmark datasets: Toronto3D and SensatUrban. Especially on the city-scale SensatUrban dataset, our results surpassed the current best result by 2.1\% in overall mIoU and yielded an improvement of 15.9\% on average for small-sample object categories comprising less than 2\% of the total samples, in comparison to the baseline method.