Re-Densification Meets Cross-Scale Propagation: Real-Time Compression of LiDAR Point Clouds

2025-08-28

AI Summary

LiDAR point clouds exhibit extreme geometric sparsity, rendering existing octree- or voxel-based sparse predictive coding methods inefficient in contextual modeling and limiting both compression performance and speed. To address this, we propose a geometry-aware densification and cross-scale feature propagation framework for real-time compression. First, a lightweight geometry re-densification module extracts robust sparse geometric features at a denser scale. Second, a multi-scale occupancy-guided cross-scale feature propagation mechanism enhances contextual modeling while suppressing redundant computation. Integrated with octree representation and hierarchical predictive coding, the framework significantly improves encoding/decoding efficiency. Evaluated on the KITTI dataset, our method achieves state-of-the-art compression ratios and attains 26 FPS for both encoding and decoding under 12-bit quantization—marking the first demonstration of high-fidelity, real-time LiDAR point cloud compression.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

LiDAR point clouds are fundamental to various applications, yet high-precision scans incur substantial storage and transmission overhead. Existing methods typically convert unordered points into hierarchical octree or voxel structures for dense-to-sparse predictive coding. However, the extreme sparsity of geometric details hinders efficient context modeling, thereby limiting their compression performance and speed. To address this challenge, we propose to generate compact features for efficient predictive coding. Our framework comprises two lightweight modules. First, the Geometry Re-Densification Module re-densifies encoded sparse geometry, extracts features at denser scale, and then re-sparsifies the features for predictive coding. This module avoids costly computation on highly sparse details while maintaining a lightweight prediction head. Second, the Cross-scale Feature Propagation Module leverages occupancy cues from multiple resolution levels to guide hierarchical feature propagation. This design facilitates information sharing across scales, thereby reducing redundant feature extraction and providing enriched features for the Geometry Re-Densification Module. By integrating these two modules, our method yields a compact feature representation that provides efficient context modeling and accelerates the coding process. Experiments on the KITTI dataset demonstrate state-of-the-art compression ratios and real-time performance, achieving 26 FPS for both encoding and decoding at 12-bit quantization. Code is available at https://github.com/pengpeng-yu/FastPCC.

Problem

• Real-time compression of LiDAR point clouds
• Addressing extreme sparsity hindering efficient context modeling
• Reducing storage and transmission overhead of high-precision scans

Innovation

• Geometry Re-Densification for sparse geometry
• Cross-scale Feature Propagation across resolutions
• Compact feature representation for efficient coding