Deep Implicit Volume Compression

TL;DR

This paper introduces a neural network-based method for compressing 3D volumetric data and textures, achieving significant bitrate reduction while maintaining high reconstruction quality, suitable for efficient storage and transmission of 3D scenes.

Contribution

It presents a novel end-to-end neural compression approach for TSDF voxel grids and textures, with lossless sign compression to prevent topological errors and a fast UV parameterization for textures.

Findings

Reduced bitrate by 66% at the same distortion

Achieved 50% lower distortion at the same bitrate

Demonstrated effectiveness on 4D performance capture datasets

Abstract

We describe a novel approach for compressing truncated signed distance fields (TSDF) stored in 3D voxel grids, and their corresponding textures. To compress the TSDF, our method relies on a block-based neural network architecture trained end-to-end, achieving state-of-the-art rate-distortion trade-off. To prevent topological errors, we losslessly compress the signs of the TSDF, which also upper bounds the reconstruction error by the voxel size. To compress the corresponding texture, we designed a fast block-based UV parameterization, generating coherent texture maps that can be effectively compressed using existing video compression algorithms. We demonstrate the performance of our algorithms on two 4D performance capture datasets, reducing bitrate by 66% for the same distortion, or alternatively reducing the distortion by 50% for the same bitrate, compared to the state-of-the-art.