Multiscale deep context modeling for lossless point cloud geometry compression

TL;DR

This paper introduces MSVoxelDNN, a multiscale deep generative model for lossless point cloud geometry compression that achieves faster encoding/decoding and 17.5% average rate savings over MPEG G-PCC.

Contribution

The paper presents a novel multiscale architecture for deep point cloud compression that improves speed and compression efficiency over previous autoregressive models.

Findings

Speeds up encoding and decoding times significantly.

Achieves 17.5% average rate savings over G-PCC.

Outperforms previous VoxelDNN in efficiency.

Abstract

We propose a practical deep generative approach for lossless point cloud geometry compression, called MSVoxelDNN, and show that it significantly reduces the rate compared to the MPEG G-PCC codec. Our previous work based on autoregressive models (VoxelDNN) has a fast training phase, however, inference is slow as the occupancy probabilities are predicted sequentially, voxel by voxel. In this work, we employ a multiscale architecture which models voxel occupancy in coarse-to-fine order. At each scale, MSVoxelDNN divides voxels into eight conditionally independent groups, thus requiring a single network evaluation per group instead of one per voxel. We evaluate the performance of MSVoxelDNN on a set of point clouds from Microsoft Voxelized Upper Bodies (MVUB) and MPEG, showing that the current method speeds up encoding/decoding times significantly compared to the previous VoxelDNN, while having average rate saving over G-PCC of 17.5%. The implementation is available at https://github.com/Weafre/MSVoxelDNN.