Multiscale Point Cloud Geometry Compression

TL;DR

This paper introduces a multiscale deep learning framework for compressing 3D point cloud geometry efficiently by leveraging sparsity, outperforming existing standards in compression rate and runtime.

Contribution

The paper presents a novel hierarchical autoencoder-based approach for point cloud compression that combines lossless octree encoding with learned probabilistic feature compression.

Findings

Achieves over 40% BD-Rate reduction compared to V-PCC.

Achieves over 70% BD-Rate reduction compared to G-PCC.

Encoding runtime is comparable to G-PCC, significantly faster than V-PCC.

Abstract

Recent years have witnessed the growth of point cloud based applications because of its realistic and fine-grained representation of 3D objects and scenes. However, it is a challenging problem to compress sparse, unstructured, and high-precision 3D points for efficient communication. In this paper, leveraging the sparsity nature of point cloud, we propose a multiscale end-to-end learning framework which hierarchically reconstructs the 3D Point Cloud Geometry (PCG) via progressive re-sampling. The framework is developed on top of a sparse convolution based autoencoder for point cloud compression and reconstruction. For the input PCG which has only the binary occupancy attribute, our framework translates it to a downscaled point cloud at the bottleneck layer which possesses both geometry and associated feature attributes. Then, the geometric occupancy is losslessly compressed using an octree codec and the feature attributes are lossy compressed using a learned probabilistic context model.Compared to state-of-the-art Video-based Point Cloud Compression (V-PCC) and Geometry-based PCC (G-PCC) schemes standardized by the Moving Picture Experts Group (MPEG), our method achieves more than 40% and 70% BD-Rate (Bjontegaard Delta Rate) reduction, respectively. Its encoding runtime is comparable to that of G-PCC, which is only 1.5% of V-PCC.