Ghost-FWL: A Large-Scale Full-Waveform LiDAR Dataset for Ghost Detection and Removal

2026-03-30

AI Summary

This study addresses the severe degradation of 3D mapping and localization caused by ghost points generated from highly reflective surfaces such as glass in mobile LiDAR systems. To tackle this challenge, we introduce the first ghost detection task tailored for full-waveform LiDAR (FWL) in dynamic mobile scenarios. We present Ghost-FWL, the largest annotated FWL dataset to date, and propose FWL-MAE, a mask autoencoder-based self-supervised representation learning method that effectively discriminates true echoes from ghost artifacts in sparse, dynamic data. Experimental results demonstrate that our approach substantially outperforms existing methods, reducing LiDAR SLAM trajectory error by 66% and decreasing false positive rates in 3D object detection by a factor of 50.

Abstract

LiDAR has become an essential sensing modality in autonomous driving, robotics, and smart-city applications. However, ghost points (or ghosts), which are false reflections caused by multi-path laser returns from glass and reflective surfaces, severely degrade 3D mapping and localization accuracy. Prior ghost removal relies on geometric consistency in dense point clouds, failing on mobile LiDAR's sparse, dynamic data. We address this by exploiting full-waveform LiDAR (FWL), which captures complete temporal intensity profiles rather than just peak distances, providing crucial cues for distinguishing ghosts from genuine reflections in mobile scenarios. As this is a new task, we present Ghost-FWL, the first and largest annotated mobile FWL dataset for ghost detection and removal. Ghost-FWL comprises 24K frames across 10 diverse scenes with 7.5 billion peak-level annotations, which is 100x larger than existing annotated FWL datasets. Benefiting from this large-scale dataset, we establish a FWL-based baseline model for ghost detection and propose FWL-MAE, a masked autoencoder for efficient self-supervised representation learning on FWL data. Experiments show that our baseline outperforms existing methods in ghost removal accuracy, and our ghost removal further enhances downstream tasks such as LiDAR-based SLAM (66% trajectory error reduction) and 3D object detection (50x false positive reduction). The dataset and code is publicly available and can be accessed via the project page: https://keio-csg.github.io/Ghost-FWL

Problem

• ghost points
• LiDAR
• multi-path reflections
• full-waveform LiDAR
• mobile sensing

Innovation

• Full-Waveform LiDAR
• Ghost Point Removal
• Self-Supervised Learning
• Masked Autoencoder
• Mobile LiDAR Dataset