A novel density-based representation for point cloud and its ability to facilitate classification

Currently, in the field of processing 3D point cloud data, two primary representation methods have emerged: point-based methods and voxel-based methods. However, the former suffer from significant computational costs and lack the ease of handling exhibited by voxel-based methods. Conversely, the later often encounter challenges related to information loss resulting from downsampling operations, thereby impeding subsequent tasks. To address these limitations, this article introduces a novel density-based representation method for voxel partitioning. Additionally, a corresponding network structure is devised to extract features from this specific density representation, thereby facilitating the successful completion of classification tasks. The experiments are implemented on ModelNet40 and MNIST demonstrate that the proposed 3D convolution can achieve the-state-of-the-art performance based on the voxels. We analyze the differences between the common ideas of convolution on the raw point clouds and the 3D convolution on the voxels. Dedicated algorithms are designed to facilitate the processing for the feature extraction on the proposed density-based representation. Furthermore, a density-based convolutional network is proposed to tackle the particular form of representation. image