Three-Dimensional Point Cloud Segmentation Algorithm Based on Depth Camera for Large Size Model Point Cloud Unsupervised Class Segmentation.

This paper proposes a 3D point cloud segmentation algorithm based on a depth camera for large-scale model point cloud unsupervised class segmentation. The algorithm utilizes depth information obtained from a depth camera and a voxelization technique to reduce the size of the point cloud, and then uses clustering methods to segment the voxels based on their density and distance to the camera. Experimental results show that the proposed algorithm achieves high segmentation accuracy and fast segmentation speed on various large-scale model point clouds. Compared with recent similar works, the algorithm demonstrates superior performance in terms of accuracy metrics, with an average Intersection over Union (IoU) of 90.2% on our own benchmark dataset.

Method for measuring the center of mass and moment of inertia of a model using 3D point clouds.

We propose a method for measuring the center of mass and moment of inertia of a model using 3D point clouds. First, we use point cloud registration and segmentation to obtain point clouds of model parts with different material densities. Then, we correct the point cloud coordinates by principal component analysis, and we perform 2.5D volume calculations in three directions. Finally, we input the material density for each model part, perform point cloud reconstruction, and calculate the center of mass and moment of inertia. In experiments, we measured the center of mass and moment of inertia simultaneously by using multiple time-of-flight cameras. The results demonstrated that the proposed method accurately measured the center of mass and moment of inertia with errors of <1m m and <3%, respectively. This is a promising approach for automatically measuring models of aircraft, cars, etc.