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

Kazuma Ikeda; Ryosei Hara; Rokuto Nagata; Ozora Sako. Zihao Ding; Takahiro Kado; Ibuki Fujioka; Taro Beppu; Mariko Isogawa

cs.CVMar 30, 2026

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

Authors: Kazuma Ikeda, Ryosei Hara, Rokuto Nagata, Ozora Sako. Zihao Ding, Takahiro Kado, Ibuki Fujioka, Taro Beppu, Mariko Isogawa,

Organizations: Keio University · Sony Semiconductor Solutions

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

Figures & tables

	Access & Platform			Sensor			Labels
Dataset	Year	Public	Platform	FWL	LiDAR Dim.	Ray Den.	Ghost	FWL Data	Frames/ Scenes †	Annotated Peaks
UNIST [ 52 ]	2017	✓	Stationary	✗	3D	278	✓	✗	–	–
Leddar PixSet [ 9 ]	2021	✓	Mobile	✓	3D	0.267	✗	✓	–	–
Lee et al. [ 23 ]	2023	✗	Stationary	✗	3D	278	✗	✗	–	–
FRACTAL [ 13 ]	2024	✓	Aerial	✗	2D	–	✗	✗	–	–
Scheuble et al. [ 38 ]	2025	✗	Mobile	✓	3D	2.56	✗	✓	0.24k / 2	NA

Table 1: Comparison of LiDAR real-world datasets for ghost detection and/or full-waveform analysis. Our Ghost-FWL contains mobile LiDAR full-waveform measurements and is one hundred times larger than prior work, making it the largest annotated FWL dataset.

Figure 2: Overview of the Ghost-FWL. Left: Our dataset includes both indoor and outdoor scenes. Based on the dense 3D maps as shown in Scene, we annotated FWL data with semantic labels: Ghost (red), Object (green), Glass (blue), Noise . Gray regions are excluded from annotation. Right shows the data acquisition setup and dataset statistics, including the incident angle distribution and LiDAR positions examples. Data were collected at three different times of day: Morning (AM10–12), Daytime (PM12–5), and Evening (PM5–7).

Figure 3: FWL-based ghost removal framework. Given FWL data, our framework predicts and removes ghost-related signals. Our model consists of a transformer-based encoder and an MLP head. We further introduce FWL-MAE, a masked autoencoder designed for representation learning on FWL data, explicitly trained to reconstruct peak position , amplitude and width . The ghosts detected by our model are then removed from FWL data, and the cleaned data are utilized for downstream tasks such as SLAM and 3D object detection.

Figure 4: Peak classification results and point cloud visualization after applying ghost removal. All results were obtained using the proposed framework. Red, green and blue indicates Ghost , Object , Glass , respectively.

Method	Recall ( $\uparrow$ )	Removal Rate ( $\uparrow$ )
MARMOT [ 39 ]	0.746	0.910
Ours w/o FWL-MAE	0.704	0.900
Ours	0.751	0.918

Table 2: Comparison of ghost removal performance with other methods.

Method	ATE [m] ( $\downarrow$ )	RTE [m] ( $\downarrow$ )
Dual-Peak [ 32 , 17 ]	0.715 $\pm$ 0.433	0.741 $\pm$ 0.406
Multi-Peak [ 27 ]	1.547 $\pm$ 1.394	1.602 $\pm$ 1.381
Ours	0.245 $\pm$ 0.138	0.245 $\pm$ 0.131

Table 3: SLAM performance under different LiDAR signal preprocessing methods.

Figure 5: Trajectory and mapping generated by SLAM using Multi-Peak processing (left) and our ghost removal method (right). Multi-Peak processing includes numerous ghost points in the reconstructed map, leading to trajectory drift. The proposed method yields a trajectory that more closely follows the ground-truth path (white) by effectively removing ghost artifacts.

Method	Ghost FP Rate [%] ( $\downarrow$ )
Dual-Peak [ 32 , 17 ]	75.8
Multi-Peak [ 27 ]	67.9
Ours	1.34

Table 4: Ghost-induced false positives in ghost detection.

Figure 6: Qualitative evaluation of 3D object detection with Multi-Peak processing (left) and our ghost removal (right). Green bounding boxes indicate persons. With Multi-Peak, a ghost person is detected behind the glass wall, whereas our method suppresses this false detection.

Appendix figures & tables16 assets

Supplementary material from the paper’s appendix.

Appendix

Scene	Frames	Location
001	2500	Indoor
002	2500	Indoor
003	2749	Indoor
004	1853	Indoor
005	2500	Outdoor
006	2445	Outdoor

Appendix

Table 5: Ghost-FWL statistics by scene.

Figure 7: All scenes in Ghost-FWL. Our dataset includes both indoor and outdoor scenes. Based on the dense 3D maps as shown in Scene, we annotated FWL data with semantic labels: Ghost (red), Object (green), Glass (blue), Noise . Gray regions are excluded from annotation. Red crosses indicate the LiDAR positions during data acquisition. The RGB images show the scenery of the capture locations.

Scene	Threshold $\tau$	# Glass Areas
001	0.5	8
002	0.5	48
003	0.5	10
004	0.5	9
005	0.5	15
006	0.5	5

Appendix

Table 6: Annotation parameters for each scene in the Ghost-FWL dataset. The threshold

\tau

is the nearest-neighbor distance used to distinguish Object and Ghost points, and “# Glass Areas” denotes the number of manually annotated glass regions

\mathcal{G}

Figure 8: Example of LiDAR position and perspective.

Figure 9: Example of 1 frame FWL data(top) and 50

\times

accumulation FWL data (bottom).

Figure 10: Example of raw annotation (top) and processed annotation for training (bottom). The circles indicate the annotation points for each class: Object (green), Glass (blue) and Ghost (red).

Name	Layer setting	Output dimension
Patch Embedding	3D Conv	$B\times N_{\text{patch}}\times D_{\text{encoder}}$
Patch Embedding	Flatten + Transpose	$B\times N_{\text{patch}}\times D_{\text{encoder}}$
Positional Encoding	Sinusoidal Position Encoding (PE)	$B\times N_{\text{patch}}\times D_{\text{encoder}}$
Masking	Select Unmasked Tokens + PE	$B\times N_{\text{unmasked}}\times D_{\text{encoder}}$
Encoder	Transformer Blocks $\times$ 6	(pretraining) $B\times N_{\text{unmasked}}\times D_{\text{encoder}}$
Encoder	LayerNorm	(fine-tuning) $B\times N_{\text{patch}}\times D_{\text{encoder}}$

Appendix

Table 7: FWL-MAE Architecture.

Method	Recall ( $\uparrow$ )
3D U-Net [ 8 ]	0.391
Lindell et al. [ 25 ]	0.641
MARMOT [ 39 ]	0.746
Ours w/o FWL-MAE	0.704
Ours	0.751

Appendix

Table 8: Comparison of ghost removal performance with other methods.

Figure 11: Peak classification results. Red, green and blue indicates Ghost , Object and Glass , respectively.

Train Data	Method	Recall ( $\uparrow$ )
100%	Ours w/o FWL-MAE	0.704
	Ours	0.751 (+ 0.047)
70%	Ours w/o FWL-MAE	0.602
	Ours	0.692 (+ 0.090)
50%	Ours w/o FWL-MAE	0.403
	Ours	0.603 (+ 0.200)

Appendix

Table 9: Ablation study of FWL-MAE.

Method	Params [M]	FLOPs [G]	FPS †	FPS ‡
3D U-Net [ 8 ]	90.3	7610	6.6	0.45
Lindell et al. [ 25 ]	1.8	4090	7.0	0.48
Ours	194.1	23.5	32.1	2.35

Appendix

Table 10: Computational cost and inference time. FPS † denotes the inference speed for model input size

(H,W,T)=(128,128,256)

. FPS ‡ denotes the inference for full-frame size

(H,W,T)=(332,400,256)

Figure 12: Our model’s ghost detection on non-glass surfaces

	Raw	Statistical Outlier Filter	Radius-based Outlier Filter
Dual-Peak ATE [m] ( $\downarrow$ )	1.248 $\pm$ 0.947	0.639 $\pm$ 0.573	0.503 $\pm$ 0.593
Multi-Peak ATE [m] ( $\downarrow$ )	2.489 $\pm$ 1.432	1.232 $\pm$ 0.928	0.887 $\pm$ 1.079
Ours ATE [m] ( $\downarrow$ )	0.294 $\pm$ 0.244	0.248 $\pm$ 0.171	0.328 $\pm$ 0.303
Dual-Peak RTE [m] ( $\downarrow$ )	1.221 $\pm$ 0.928	0.608 $\pm$ 0.529	0.471 $\pm$ 0.555
Multi-Peak RTE [m] ( $\downarrow$ )	2.513 $\pm$ 1.440	1.280 $\pm$ 0.901	0.832 $\pm$ 1.015
Ours RTE [m] ( $\downarrow$ )	0.288 $\pm$ 0.237	0.243 $\pm$ 0.157	0.319 $\pm$ 0.291

Appendix

Table 11: SLAM performance ablation with different point-cloud processing methods.

Figure 13: Trajectory and mapping results using Dual-Peak processing (left), Multi-Peak processing (center), and our ghost-removal method (right). The bottom image shows the scenery of the SLAM evaluation, recorded in a corridor enclosed by glass railings and doors. The scene is same to that used in the main paper Fig 5 , with the RGB scene image and the visualization results for the Dual-Peak method additionally included.

Scene	Indoor/ Outdoor	Pedestrians	Frames
PedScene 001	Indoor	2	14
PedScene 002	Outdoor	3	31
PedScene 003	Indoor	3	32
PedScene 004	Indoor	1	25
TOTAL	Indoor 3 / Outdoor 1	1 / 2 / 3	102

Appendix

Table 12: Details of the object-detection test dataset (PedScene).

Figure 14: Example scenes from the PedScene dataset used for 3D object detection. The dataset includes both indoor and outdoor environments with glass surfaces, building entrances, and glass-walled sidewalks.

Point Cloud Anomaly Detection

Point cloud anomaly detection identifies and localizes unusual 3D geometry or surface defects, often in industrial inspection with few labeled failures.

See all 25 papers

GhostPoint: Self-Supervised Representation Learning by Hallucinating Occluded LiDAR Structure
Aug 14, 2026Mohamed Abdelsamad, Bin Yang, Michael Ulrich +4Self-Supervised Point Cloud Learning Lidar-Based 3D Object Detection
Bosch Center for AI · University of Freiburg · University of Luebeck
Ghosts in the Point Clouds: De-glaring LiDAR in the Transient Domain
May 23, 2026Avery Gump, Connor Henley, Sungjin Cheong +2
University of Wisconsin–Madison, Madison, WI, USA
AevaScenes: An FMCW LiDAR Dataset and Benchmark for Long-Range Perception
Sep 27, 2026Gautham Narayan Narasimhan, Heethesh Vhavle, Kumar Bhargav Viswanatha +2Scene Flow Estimation Lidar-Based 3D Object Detection
Aeva Inc., Mountain View, CA, USA. · Carnegie Mellon University, Pittsburgh, PA, USA.

Aug 14, 2026J/K move · Enter open · S save

GhostPoint: Self-Supervised Representation Learning by Hallucinating Occluded LiDAR Structure

Mohamed Abdelsamad, Bin Yang, Michael Ulrich, Miao Zhang+3

Bosch Center for AI · University of Freiburg · University of Luebeck

Self-Supervised Point Cloud Learning Lidar-Based 3D Object Detection Point Cloud Learning 3D Representation Learning Self-Supervised Visual Representation Learning

Open paper AskPDF arXiv

3D object detection from LiDAR point clouds is a core problem in autonomous driving. Recent advances in self-supervised learning (SSL) enable scalable pretraining and transfers well to per-point tasks such as semantic and panoptic segmentation, but transfer to 3D detection remains weaker. We analyze recent SSL methods and find that most objectives are defined only on measured LiDAR returns from visible surfaces, leaving occluded and unobserved regions unconstrained. This visible-surface bias can be sufficient for point-wise prediction, but 3D detection requires robustness to missing structure. To address this gap, we propose GhostPoint, an SSL framework that hallucinates latent features in local neighborhoods around discovered instances, generated via a novel instance voxel dilation. In GhostPoint, an encoder processes observed returns, and an additional predictor infers neighborhood representations from observed context. In addition to standard encoder-level supervision, we introduce a predictor-level supervision scheme on sampled voxels from generated neighborhoods. Specifically, observed (visible/masked) voxels match teacher-encoder targets, while unobserved voxels match teacher-predictor hallucinations. This design encourages the learned representation to explicitly model structure beyond observed returns. Extensive evaluations on nuScenes and Waymo demonstrate that our method achieves state-of-the-art performance, consistently improving downstream 3D detection, especially under sparse scans and limited labels.

Abstract

Point Cloud Anomaly Detection

Explore similar work

GhostPoint: Self-Supervised Representation Learning by Hallucinating Occluded LiDAR Structure

Ghosts in the Point Clouds: De-glaring LiDAR in the Transient Domain

AevaScenes: An FMCW LiDAR Dataset and Benchmark for Long-Range Perception