Spherical Interpolated Convolutional Network with Distance-Feature Density for 3D Semantic Segmentation of Point Clouds
This work aims to improve the accuracy and efficiency of 3D semantic segmentation for robots by developing a more suitable convolution operator for unstructured point clouds, which is an incremental improvement for the robotics and computer vision communities.
This paper addresses the challenge of 3D semantic segmentation of unstructured point clouds by proposing a spherical interpolated convolution operator to replace traditional grid-shaped 3D convolution. The new operator improves network accuracy and reduces parameters, while a self-learned distance-feature density method enhances feature extraction rationality and effectiveness.
The semantic segmentation of point clouds is an important part of the environment perception for robots. However, it is difficult to directly adopt the traditional 3D convolution kernel to extract features from raw 3D point clouds because of the unstructured property of point clouds. In this paper, a spherical interpolated convolution operator is proposed to replace the traditional grid-shaped 3D convolution operator. This newly proposed feature extraction operator improves the accuracy of the network and reduces the parameters of the network. In addition, this paper analyzes the defect of point cloud interpolation methods based on the distance as the interpolation weight and proposes the self-learned distance-feature density by combining the distance and the feature correlation. The proposed method makes the feature extraction of spherical interpolated convolution network more rational and effective. The effectiveness of the proposed network is demonstrated on the 3D semantic segmentation task of point clouds. Experiments show that the proposed method achieves good performance on the ScanNet dataset and Paris-Lille-3D dataset.