Preprocessing and Transmission for 3D Point Cloud Data

Robots play an increasingly important role in social life, especially the front desk robots. But the front desk robots seldom handle business in reality unless they have proper functionalities for Human-Robot Interaction (HRI). For enhancing the immersed sense in the interactive process, we consider the 3D image of the upper body of the operator in the remote control room as the upper body of front desk robots. However, it is a great challenge to transmit the 3D point cloud data in the way of remote interaction. The paper uses a simple method to deal with the problem of transmitting the 3D point cloud data and the idea of the method is to reduce the network data volume. In order to reduce the network data volume, we only consider that the 3D point cloud of interest will be transmitted. The filters are used to remove the noise and background of the 3D image, the segmentation algorithm will be used to acquire the 3D point cloud data of interest. The experiment result demonstrates that the method can reduce the network data volume and ensure high-quality image information. Thus, the method can reduce transmission time.