LiZIP: An Auto-Regressive Compression Framework for LiDAR Point Clouds

2026-03-24

AI Summary

This work addresses the challenges of real-time processing and V2X transmission bottlenecks in autonomous driving caused by the large volume of LiDAR point cloud data by proposing a lightweight, near-lossless, and drift-free compression framework. Leveraging local context, the method employs a compact multilayer perceptron (MLP) for autoregressive coordinate prediction and encodes only sparse residuals via neural prediction, achieving both low computational overhead and high compression efficiency. Evaluated on the NuScenes and Argoverse datasets, the approach reduces file size by 7.5%–14.8% compared to LASzip, outperforms Google Draco by 8.8%–11.3%, and improves compression efficiency by 38%–48% over GZip, while demonstrating strong cross-dataset generalization capability.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Vision for Robotics & Autonomous Driving
• Data Mining & Knowledge Management: Data Compression

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

The massive volume of data generated by LiDAR sensors in autonomous vehicles creates a bottleneck for real-time processing and vehicle-to-everything (V2X) transmission. Existing lossless compression methods often force a trade-off: industry standard algorithms (e.g., LASzip) lack adaptability, while deep learning approaches suffer from prohibitive computational costs. This paper proposes LiZIP, a lightweight, near-lossless zero-drift compression framework based on neural predictive coding. By utilizing a compact Multi-Layer Perceptron (MLP) to predict point coordinates from local context, LiZIP efficiently encodes only the sparse residuals. We evaluate LiZIP on the NuScenes and Argoverse datasets, benchmarking against GZip, LASzip, and Google Draco (configured with 24-bit quantization to serve as a high-precision geometric baseline). Results demonstrate that LiZIP consistently achieves superior compression ratios across varying environments. The proposed system achieves a 7.5%-14.8% reduction in file size compared to the industry-standard LASzip and outperforms Google Draco by 8.8%-11.3% across diverse datasets. Furthermore, the system demonstrates generalization capabilities on the unseen Argoverse dataset without retraining. Against the general purpose GZip algorithm, LiZIP achieves a reduction of 38%-48%. This efficiency offers a distinct advantage for bandwidth constrained V2X applications and large scale cloud archival.

Problem

• LiDAR point cloud compression
• real-time processing
• V2X transmission
• lossless compression
• computational efficiency

Innovation

• neural predictive coding
• lightweight compression
• zero-drift
• point cloud compression
• MLP-based prediction