U-Motion: Learned Point Cloud Video Compression with U-Structured Temporal Context Generation

TL;DR

U-Motion introduces a novel learning-based point cloud video compression method utilizing a U-Structured temporal context generation framework, significantly improving compression efficiency for geometry and attributes.

Contribution

The paper presents a U-Structured inter-frame prediction framework with multi-scale motion estimation and compensation, advancing point cloud video compression techniques.

Findings

Achieves significant compression gains over MPEG G-PCC-GesTM v3.0.

Outperforms recent learning-based methods in geometry and attribute compression.

Demonstrates effectiveness on dense dynamic point clouds.

Abstract

Point cloud video (PCV) is a versatile 3D representation of dynamic scenes with emerging applications. This paper introduces U-Motion, a learning-based compression scheme for both PCV geometry and attributes. We propose a U-Structured inter-frame prediction framework, U-Inter, which performs explicit motion estimation and compensation (ME/MC) at different scales with varying levels of detail. It integrates Top-Down (Fine-to-Coarse) Motion Propagation, Bottom-Up Motion Predictive Coding and Multi-scale Group Motion Compensation to enable accurate motion estimation and efficient motion compression at each scale. In addition, we design a multi-scale spatial-temporal predictive coding module to capture the cross-scale spatial redundancy remaining after U-Inter prediction. We conduct experiments following the MPEG Common Test Condition for dense dynamic point clouds and demonstrate that U-Motion can achieve significant gains over MPEG G-PCC-GesTM v3.0 and recently published learning-based methods for both geometry and attribute compression.