Automotive
Radar Ghost
Radar multipath perception
Stationary automotive radar recordings with semantic instance labels for real and ghost reflections.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- Automotive
- Environment
- Outdoor
- Data origin
- Real-world
- Sensor count
- 3
- Ground truth
- Not reported
- Calibration
- Not reported
- Annotations
- Available
Sensor overview
High-level sensor availability before the detailed sensor records below.
Ground truth
Reference scope, method, rate, coverage, and provenance.
Not reported
Calibration and synchronization
Reported calibration level, reproducibility signals, and supporting notes.
- Level
- Not reported
- Processed parameters
- Not reported
- Raw calibration data
- Not reported
- Calibration targets
- Not reported
Calibration data
Calibration reproducibility details are not reported.
Notes
Not reportedKnown or intentional limitations
Reported constraints and characteristics to check before using the dataset.
The original collection has 111 recordings across 21 controlled scenarios with a stationary ego vehicle. Virtual multi-object overlays combine real recordings and are not additional independent captures in this count.
Use v1.1 for LiDAR: it fixes a v1.0 radar/LiDAR timestamp issue, removes noise and changes LiDAR coordinate fields. Older parsers may require updates.
Five road-user classes are separate from reflection order/type and ignore/sketchy labels. Documentation photographs do not establish continuous camera streams.
No independent platform trajectory, raw GNSS stream, universally achieved timing bound or complete raw calibration release is established.
Cameras
Not reported
IMUs
Not reported
GNSS
Not reported
LiDAR
LiDAR 1
- Model or name
- LiDAR 1
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- 4
- Range
- Not reported
Radar
Experimental automotive radars
- Count
- 2
- Released data
- Point detections / point clouds
- Frame rate
- 10 Hz
- Doppler velocity
- Not reported
- Elevation angle
- Not reported
- Configured range
- Not reported
- Field of view
- Not reported
- Notes
- Point detections; raw radar cubes are not established as released.
Additional sensors
Not reported
Annotations
Reported annotation availability and task support.
- Availability
- Available
- Format
- Point-level road-user semantics and instances, with separate multipath-type/order and ignore/sketchy labels in HDF5.
- Class count
- 5
Semantic segmentation
Available
Instance segmentation
Available
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.
@inproceedings{kraus2021radarghost,
author = {Florian Kraus and Nicolas Scheiner and Werner Ritter and Klaus Dietmayer},
title = {The Radar Ghost Dataset – An Evaluation of Ghost Objects in Automotive Radar Data},
booktitle = {2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
year = {2021},
pages = {8570--8577},
doi = {10.1109/IROS51168.2021.9636338},
url = {https://arxiv.org/abs/2404.01437v1}
}Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy