Pandar128 dataset for lane line detection

TL;DR

The paper introduces Pandar128, a large-scale LiDAR dataset for lane detection, along with a simple baseline method and a new evaluation metric, to advance research in autonomous driving under diverse conditions.

Contribution

It provides the largest public dataset for LiDAR-based lane detection, a lightweight baseline method, and a novel evaluation metric to standardize performance assessment.

Findings

SimpleLidarLane achieves strong results in challenging conditions.

The dataset covers diverse real-world scenarios in Germany.

The new IAM-F1 metric improves evaluation consistency.

Abstract

We present Pandar128, the largest public dataset for lane line detection using a 128-beam LiDAR. It contains over 52,000 camera frames and 34,000 LiDAR scans, captured in diverse real-world conditions in Germany. The dataset includes full sensor calibration (intrinsics, extrinsics) and synchronized odometry, supporting tasks such as projection, fusion, and temporal modeling. To complement the dataset, we also introduce SimpleLidarLane, a light-weight baseline method for lane line reconstruction that combines BEV segmentation, clustering, and polyline fitting. Despite its simplicity, our method achieves strong performance under challenging various conditions (e.g., rain, sparse returns), showing that modular pipelines paired with high-quality data and principled evaluation can compete with more complex approaches. Furthermore, to address the lack of standardized evaluation, we propose a novel polyline-based metric - Interpolation-Aware Matching F1 (IAM-F1) - that employs interpolation-aware lateral matching in BEV space. All data and code are publicly released to support reproducibility in LiDAR-based lane detection.