PointCompress3D - A Point Cloud Compression Framework for Roadside LiDARs in Intelligent Transportation Systems

title_figure
Visualization of the processing pipeline of our PointCompress3D point cloud compression and streaming framework. Our framework takes raw roadside LiDAR point clouds as input, processes them, and outputs compressed point clouds to facilitate downstream application tasks like 3D object detection, data storage, and real-time point cloud streaming on an ITS test bed for autonomous driving.

Overview

PointCompress3D is the first point cloud compression framework for roadside LiDARs. With this framework you can select a point cloud compression algorithm to store roadside point clouds efficiently and stream them in real-time using the state-of-the-art loosy and lossless compression algorithms.

In summary:
  • We propose a point cloud compression framework for roadside infrastructure LiDAR sensors.
  • We provide an in-depth comparison of state-of-the-art compression methods on the SemanticKITTI, Ford and TUMTraf dataset family.
  • We extend existing compression methods to make them compatible with our roadside Ouster LiDAR sensors.
  • We perform extensive experiments and ablation studies on the TUMTraf Intersection and TUMTraf V2X Coop. Perception dataset.
  • We open-source our framework, which contains the point cloud projection and compression module and provide a project website with video results.

Abstract

In the context of Intelligent Transportation Systems (ITS), efficient data compression is crucial for managing large-scale point cloud data acquired by roadside LiDAR sensors. The demand for efficient storage, streaming, and real-time object detection capabilities for point cloud data is substantial. This work introduces PointCompress3D, a novel point cloud compression framework explicitly tailored for roadside LiDARs. Our framework addresses the challenges of compressing high-resolution point clouds in real time while maintaining accuracy and compatibility with roadside LiDAR sensors. We adapt, extend, and integrate three cutting-edge compression methods and evaluate them on the real-world TUM Traffic datasets. Moreover, we deploy our compression framework on a real ITS test bed for autonomous driving and test it under real traffic conditions. After fine-tuning, we achieve a frame rate of 10 FPS while keeping compression sizes below 105 Kb, a reduction of 50 times, and maintaining object detection performance on par with the original data. In extensive experiments and ablation studies, we finally achieved a PSNR d2 of 94.46 and a BPP of 6.54 on the TUM Traffic datasets. The code is available on our project website: https://pointcompress3d.github.io.

Qualitative Results

title_figure
Qualitative results on the TUMTraf Intersection dataset. From left to right: a) Side view of a truck and a van in the original point cloud. b) Experiment E1: Reconstructed point cloud with a subsampling distance of 3.0 and a minimum kernel radius of 1.5. c) Experiment E3: Reconstructed point cloud with a subsampling distance of 1.0 and a minimum kernel radius of 1.2. Both compressed point cloud projection images are generated with max. 30,000 points and a grid size of 40x40x15 m.
Left: Stream of original point clouds of the TUMTraf A9 Highway dataset. Right: Reconstructed stream of point clouds.

Ablation Studies

title_figure
Adjusting the minimum kernel radius parameters by 0.5 times and 2 times the original value, respectively.
title_figure
We set the kernel radius in encoding block to 10 and 0.05 (top row) while keeping the decoding block values constant at 0.05. Subsequently, we set kernel radius in the decoding block to 1 and 0.01 (bottom row) while keeping the encoding block values constant at 1.0.
title_figure
Visualization of ablation study for the max. number of points. The point cloud shows a side view of a truck and a van with 5,000, 12,500, 25,000, 50,000, 100,000, and 200,000 points.
Left: Stream of original point clouds of the TUMTraf Intersection dataset. Right: Reconstructed stream of point clouds.

Quantitative Results

Grid Size Max. Points Enc. (ms) Dec. (ms) Enc. VRAM (GB) Dec. VRAM (GB) PSNR d1 PSNR d2 BPP mAP 3D
8x8x3 50,000 220 2.7 3.9 3.9 15.32 23.68 7.48 13.32
100,000 410 2.7 6.0 4.3 21.81 30.13 11.72 17.29
200,000 520 2.8 5.5 7.1 24.18 32.88 13.57 19.39
16x16x6 50,000 230 2.6 3.0 2.7 -7.81 -0.19 5.17 12.21
100,000 350 2.5 4.3 3.7 -3.59 3.87 7.03 19.50
200,000 510 2.7 6.8 4.0 -1.46 6.31 7.73 20.91
24x24x9 50,000 240 2.5 2.2 2.7 -5.57 0.99 3.90 13.59
100,000 410 2.6 5.5 3.1 -1.87 5.50 4.94 19.75
200,000 530 2.7 6.6 3.2 -0.18 7.47 5.31 19.32
Parameter tuning for DEPOCO on the max. number of points and the grid size. We use the TUMTraf Intersection dataset to find the best parameters and evaluate the compression method on the TUMTraf V2X Cooperative Perception dataset.