Semantic Segmentation of 3D Point Cloud Based on Spatial Eight-Quadrant Kernel Convolution

Abstract

In order to deal with the problem that some existing semantic segmentation networks for 3D point clouds generally have poor performance on small objects, a Spatial Eight-Quadrant Kernel Convolution (SEQKC) algorithm is proposed to enhance the ability of the network for extracting fine-grained features from 3D point clouds. As a result, the semantic segmentation accuracy of small objects in indoor scenes can be improved. To be specific, in the spherical space of the point cloud neighborhoods, a kernel point with attached weights is constructed in each octant, the distances between the kernel point and the points in its neighborhood are calculated, and the distance and the kernel points’ weights are used together to weight the point cloud features in the neighborhood space. In this case, the relationship between points are modeled, so that the local fine-grained features of the point clouds can be extracted by the SEQKC. Based on the SEQKC, we design a downsampling module for point clouds, and embed it into classical semantic segmentation networks (PointNet++, PointSIFT and PointConv) for semantic segmentation. Experimental results on benchmark dataset ScanNet V2 show that SEQKC-based PointNet++, PointSIFT and PointConv outperform the original networks about 1.35–2.12% in terms of MIoU, and they effectively improve the semantic segmentation performance of the networks for small objects of indoor scenes, e.g., the segmentation accuracy of small object “picture” is improved from 0.70% of PointNet++ to 10.37% of SEQKC-PointNet++.

Part2Point: A Part-Oriented Point Cloud Reconstruction Framework

“…The approach involves projecting 3D mobile point clouds into a 2D representation using spherical projection on range images. Liu et al [ 52 ] introduced the spatial eight-quadrant kernel convolution (SEQKC) algorithm for enhancing 3D point cloud semantic segmentation, specifically targeting small objects in complex environments. The SEQKC algorithm improves the network’s capability to extract detailed features, resulting in higher accuracy for small objects and boundary features.…”

Segmentation of Facial Bones from Skull Point Clouds Based on Smoothed Deviation Angle

“…To explore various information in the point cloud, segmentation becomes the basic tool to better understand the represented object [18]. In general, there are two methods of point cloud segmentation, segmentation based on geometric characteristics, and segmentation based on semantic labeling [19].…”