List of Datasets

Automotive

HeLiMOS

moving object segmentation; cross-LiDAR generalization

Point-wise motion labels for four heterogeneous LiDAR streams in HeLiPR KAIST05.

Metadata: Complete Metadata source documented Released 2024

Primary places to inspect, download, read about, or reproduce the dataset.

Quick facts

Vehicle type
Automotive
Environment
Outdoor
Data origin
Real-world
Sensor count
4
Ground truth
Available
Calibration
Parameters Only
Annotations
Available

Sensor overview

High-level sensor availability before the detailed sensor records below.

Cameras

Not reported

IMUs

Not reported

GNSS

Not reported

LiDAR

4 reported

Radar

Not reported

Magnetometers

Not reported

Barometers

Not reported

UWB

Not reported

Ground truth

Reference scope, method, rate, coverage, and provenance.

Scope Method Reference system Count Rate Coverage Reference provenance
External
Measured using equipment independent of the platform’s normal onboard sensor pipeline.
Derived
Computed primarily from the dataset’s own recorded sensors.
Hybrid
Combines independent reference equipment with onboard measurements.
6DoF GNSS / INS Inherited HeLiPR INS-based pose product Not reported Not reported Unknown Reference provenance
External
Measured using equipment independent of the platform’s normal onboard sensor pipeline.
Derived
Computed primarily from the dataset’s own recorded sensors.
Hybrid
Combines independent reference equipment with onboard measurements.
Unknown
Reference notes
Maintained toolbox exports poses.txt from parent INS-based LiDAR_GT. Annotation-local pose corrections are not independent trajectory measurements; exact release-wide error and valid intervals remain unknown.

Calibration and synchronization

Reported calibration level, reproducibility signals, and supporting notes.

Level
Parameters Only
Processed parameters
Available
Raw calibration data
Not reported
Calibration targets
Not reported

Parameters only

Processed calibration parameters are reported, but raw calibration data is not reported.

Notes

Physical extrinsics are obtained from HeLiPR. Converted SemanticKITTI-style calib.txt identity matrices describe the chosen output frame, not evidence of identity physical extrinsics or raw calibration sessions.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

Current request-based distribution provides the KAIST05 IROS 2024 version with deskewed clouds and motion labels for four LiDAR types. The 12,188 labelled-cloud total has unresolved per-sensor versus combined counting semantics.

Labels encode observed motion, with unlabelled/static/dynamic IDs; they are not semantic object categories. Sequential revisits and local pose corrections affect split independence and annotation quality.

Inherited INS poses and software-aligned cloud products are documented. Full timing/reference accuracy, native sensor cadence and standalone parent camera/IMU/GNSS/FMCW auxiliary products are not established for this extension.

Four heterogeneous scanners are physical input groups; no channel count is inferred solely from a model name. Converted identity calibration matrices should not be interpreted as physical rig geometry.

Cameras

Not reported

IMUs

Not reported

GNSS

Not reported

LiDAR

Velodyne VLP16

Model or name
Velodyne VLP16
Count
1
Dimensions
3
Rate
Not reported
Channels
Not reported
Range
Not reported

Ouster OS2-128

Model or name
Ouster OS2-128
Count
1
Dimensions
3
Rate
Not reported
Channels
Not reported
Range
Not reported

Livox Avia

Model or name
Livox Avia
Count
1
Dimensions
3
Rate
Not reported
Channels
Not reported
Range
Not reported

Aeva Aeries II

Model or name
Aeva Aeries II
Count
1
Dimensions
3
Rate
Not reported
Channels
Not reported
Range
Not reported

Radar

Not reported

Additional sensors

Not reported

Automotive

HeLiPR

Road-vehicle lidar collection with four different scanners, repeated urban routes and long-term revisits of MulRan sites.

Annotations

Reported annotation availability and task support.

Availability
Available
Format
SemanticKITTI-MOS pointwise unlabeled/static/dynamic labels; deskewed scans and inherited poses.
Class count
3
Semantic segmentation Available
Instance segmentation Not reported
Object detection Not reported
Optical flow Not reported
Depth ground truth Not reported

Citation

Citation key and BibTeX kept at the end of the page for reference.

Citation key
lim2024helimos

BibTeX

@misc{lim2024helimos,
      title={HeLiMOS: A Dataset for Moving Object Segmentation in 3D Point Clouds From Heterogeneous LiDAR Sensors}, 
      author={Hyungtae Lim and Seoyeon Jang and Benedikt Mersch and Jens Behley and Hyun Myung and Cyrill Stachniss},
      year={2024},
      eprint={2408.06328},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy