MGTN: Multi-scale Graph Transformer Network for 3D Point Cloud Semantic Segmentation

The structural similarity of point clouds presents challenges in accurately recognizing and segmenting semantic information at the demarcation points of complex scenes or objects. In this study, we propose a multi-scale graph transformer network (MGTN) for 3D point cloud semantic segmentation. First, a multi-scale graph convolution (MSG-Conv) is devised to address the limitations faced by existing methods when extracting local and global features of point cloud data with varying densities simultaneously. Subsequently, we employ a graph-transformer (G-T) module to enhance edge details and spatial position information in the point cloud, thereby improving recognition accuracy for small objects and confusing elements such as columns and beams. Extensive testing on ShapeNet parts and S3DIS datasets was conducted to demonstrate the effectiveness of MGTN. Compared to the baseline network DGCNN, our proposed MGTN achieves substantial performance improvements, as evidenced by notable increases in mIoU of 1.5% and 18.5% on the ShapeNet parts and S3DIS datasets respectively. Additionally, MGTN outperforms the recent CFSA- Net by 2.3% and 3.4% on OA and mIoU respectively.

I. Introduction

In computer vision, the goal of semantic segmentation is to assign semantic labels to each point in the 3D point cloud [1]. With the widespread use of 3D perception devices such as laser scanners and depth cameras, point cloud semantic segmentation has demonstrated great potential for applications in areas such as autonomous driving [2], robot navigation [3], and architectural digital twins [4]. This has motivated researchers to work on developing efficient and accurate semantic segmentation algorithms.