A Sampling-based 3D Point Cloud Compression Algorithm for Immersive Communication
- PDF / 3,260,005 Bytes
- 10 Pages / 595.224 x 790.955 pts Page_size
- 35 Downloads / 179 Views
A Sampling-based 3D Point Cloud Compression Algorithm for Immersive Communication Hui Yuan1 · Dexiang Zhang2 · Weiwei Wang3 · Yujun Li3
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract 3D point cloud is one of the most common and basic 3D object representation model that is widely used in virtual/augmented reality applications, e.g., immersive communication. Compression of 3D point cloud is a big challenge because of its huge data volume and irregular data structure. In this paper, we propose a sampling-based compression algorithm for 3D point clouds. First, a 3D point cloud was resampled by a graph filter to obtain a subset of representative 3D points. Then, the representative points were compressed by the G-PCC (geometry-based point cloud compression) encoder software that was released by MPEG. Finally, the decoded representative points were used to reconstruct the original 3D point clouds by a CNN-based up-sampling approach. Experimental results demonstrate that a significant (73.15%) bit rate reduction can be achieved by the proposed 3D point cloud compression algorithm with minimal quality degradation of the reconstructed 3D point clouds. Keywords 3D point cloud · Compression · Sampling · Surface reconstruction
1 Introduction With the rapid development of 3D scanning technologies, 3D point clouds have become an important and popular representation for 3D objects. It has been widely used in many fields, such as Virtual/Augmented/Mixed reality (VR/AR/MR), automated driving, digital 3D cultural relics, Yujun Li
[email protected] Hui Yuan [email protected] Dexiang Zhang [email protected] Weiwei Wang [email protected] 1
Shenzhen Research Institute of Shandong University, Shenzhen, Guangdong Province, China, and the School of Control Science and Engineering, Ji’nan, Shandong Province, China
2
College of Information Science and Engineering, Qilu Normal University, No. 2 Wenbo Road, Zhangqiu District, Ji’nan, Shangdong Province, China
3
School of Information Science and Engineering, Shandong University,Qingdao, Shandong Province, China
and 3D printing. Apart from geometry information, a 3D point cloud also comprises of many other attributes, like color and normal. Processing and communication of 3D point clouds, e.g., compression [1–8], rendering [9], matching and registration [10–13], classification and segmentation [14–19], have become an important research topic for both industry and academia. Dealing with a 3D point cloud, however, is challenging. Unlike traditional 2D images, the data structure of a 3D point cloud is irregular. In addition, 3D point clouds always come with a large number of points, in a specific application, such as digital 3D cultural relics in which billions of points associated with geometry and color should be stored. Therefore, it is very critical to compress 3D point clouds efficiently for processing and transmission. In this paper, we propose an efficient compression method for the geometry of 3D point clouds. Specifically, as shown in Fig
Data Loading...
