# Learning Convolutional Transforms for Lossy Point Cloud Geometry Compression

## TL;DR

This paper introduces a novel data-driven point cloud geometry compression method using learned convolutional transforms, outperforming existing solutions in rate-distortion efficiency and maintaining high resolution at low bitrates.

## Contribution

It proposes a new convolutional transform-based compression technique for static point clouds with joint rate-distortion optimization and a binary classification decoding approach.

## Key findings

- Outperforms MPEG reference in rate-distortion with 51.5% BDBR savings
- Maintains high resolution outputs at low bitrates
- Effective for large-scale point cloud data

## Abstract

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 .