Voxel- And Graph-Based Point Cloud Segmentation of 3d Scenes Using Perceptual Grouping Laws

Abstract

ABSTRACT:Segmentation is the fundamental step for recognizing and extracting objects from point clouds of 3D scene. In this paper, we present a strategy for point cloud segmentation using voxel structure and graph-based clustering with perceptual grouping laws, which allows a learning-free and completely automatic but parametric solution for segmenting 3D point cloud. To speak precisely, two segmentation methods utilizing voxel and supervoxel structures are reported and tested. The voxel-based data structure can increase efficiency and robustness of the segmentation process, suppressing the negative effect of noise, outliers, and uneven points densities. The clustering of voxels and supervoxel is carried out using graph theory on the basis of the local contextual information, which commonly conducted utilizing merely pairwise information in conventional clustering algorithms. By the use of perceptual laws, our method conducts the segmentation in a pure geometric way avoiding the use of RGB color and intensity information, so that it can be applied to more general applications. Experiments using different datasets have demonstrated that our proposed methods can achieve good results, especially for complex scenes and nonplanar surfaces of objects. Quantitative comparisons between our methods and other representative segmentation methods also confirms the effectiveness and efficiency of our proposals.

Window Detection from UAS-Derived Photogrammetric Point Cloud Employing Density-Based Filtering and Perceptual Organization

“…The quantification of human perception in 3D is firstly addressed here, according to its claim. Furthermore, Xu et al [42] is another paper that reveals one of the first research efforts in this area. It presents a strategy for segmentation of two types of point clouds via voxel and super voxel structures.…”

Two-Layer-Graph Clustering for Real-Time 3D LiDAR Point Cloud Segmentation

“…The visit map corresponding to the node is marked as visited (line 7-10). When the node queue is not empty, take the first node of the queue and mark its category as the current category label (line [11][12][13][14]. Search the nodes near the current node (the nodes in upper and lower adjacent channels, which are connected to the current node and the first unconnected node on both sides, as shown in Figure 7).…”

“…The methods mentioned above are representative of the grid map method. Xu [11] uses a voxel and graph-based strategy to reduce point cloud density, and Jeremie [12] also reduces the number of regions by voxel. Although this kind of method can significantly reduce the running time, the accuracy of the results depends on the grid or voxel accuracy, and there are a lot of blank areas in the grid map, which waste space.…”

Virtual Object Replacement Based on Real Environments: Potential Application in Augmented Reality Systems

“…Thus, it is really only suitable for the segmentation of basic shapes, such as cylinders. Graph-based segmentation [26,27] does not require the definition of seed points and is not limited in terms of object shape. Therefore, training is not required.…”