Learning Convolutional Transforms for Lossy Point Cloud Geometry Compression
This addresses the problem of high-resolution point cloud compression for virtual and mixed reality applications, representing a strong specific gain.
The paper tackles efficient compression of static point cloud geometry for virtual and mixed reality by proposing a data-driven method using learned convolutional transforms and uniform quantization, achieving 51.5% BDBR savings on average compared to the MPEG reference solution.
Efficient point cloud compression is fundamental to enable the deployment of virtual and mixed reality applications, since the number of points to code can range in the order of millions. In this paper, we present a novel data-driven geometry compression method for static point clouds based on learned convolutional transforms and uniform quantization. We perform joint optimization of both rate and distortion using a trade-off parameter. In addition, we cast the decoding process as a binary classification of the point cloud occupancy map. Our method outperforms the MPEG reference solution in terms of rate-distortion on the Microsoft Voxelized Upper Bodies dataset with 51.5% BDBR savings on average. Moreover, while octree-based methods face exponential diminution of the number of points at low bitrates, our method still produces high resolution outputs even at low bitrates. Code and supplementary material are available at https://github.com/mauriceqch/pcc_geo_cnn .