Automotive
LiDAR-CS
Cross-sensor 3D object detection and domain adaptation
Six simulated LiDAR patterns share scanned street backgrounds and synthetic object placements, isolating sampling-pattern effects.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- Automotive
- Environment
- Outdoor
- Data origin
- Simulation
- Sensor count
- 11
- 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 84,000 simulated samples comprise six matched variants of 14,000 scenes, with corresponding foreground/background geometry and 7,000/7,000 train/test splits. They are not independent temporal recordings.
Sensor patterns are alternative configurations, not a simultaneously mounted rig. The composite Livox configuration has five Horizon units and one Tele-15; the flat interface can sum devices across alternatives.
Simulated measurements use scanned street backgrounds and CAD foregrounds. These assets do not establish an independently released physical measurement stream; origin remains Simulation. Frozen scan patterns and rendering assumptions limit real-world transfer.
Official files document binary XYZI clouds and KITTI-style object labels. Suggested semantic-label generation, platform trajectories, capture rates, raw calibration and new sensor patterns are not established release products.
Cameras
Not reported
IMUs
Not reported
GNSS
Not reported
LiDAR
Velodyne Puck 16
- Model or name
- Velodyne Puck 16
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- 16
- Range
- Not reported
Velodyne 32E
- Model or name
- Velodyne 32E
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- 32
- Range
- Not reported
Velodyne HDL-64E
- Model or name
- Velodyne HDL-64E
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- 64
- Range
- Not reported
Velodyne 128
- Model or name
- Velodyne 128
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- Not reported
- Range
- Not reported
ONCE-40 pattern
- Model or name
- ONCE-40 pattern
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- 40
- Range
- Not reported
Livox Horizon
- Model or name
- Livox Horizon
- Count
- 5
- Dimensions
- 3
- Rate
- Not reported
- Channels
- Not reported
- Range
- Not reported
Livox Tele-15
- Model or name
- Livox Tele-15
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- Not reported
- Range
- Not reported
Radar
Not reported
Additional sensors
Not reported
Annotations
Reported annotation availability and task support.
- Availability
- Available
- Format
- Foreground category and cuboid x,y,z,width,length,height,yaw
- Class count
- 5
Semantic segmentation
Not reported
Instance segmentation
Not reported
Object detection
Available
Optical flow
Not reported
Depth ground truth
Not reported
Citation
Citation key and BibTeX kept at the end of the page for reference.
@misc{fang2023lidarcs,
author = {{Jin Fang} and {Dingfu Zhou} and {Jingjing Zhao} and {Chenming Wu} and {Chulin Tang} and {Cheng-Zhong Xu} and {Liangjun Zhang}},
title = {LiDAR-CS Dataset: LiDAR Point Cloud Dataset with Cross-Sensors for 3D Object Detection},
year = {2023},
url = {https://arxiv.org/abs/2301.12515v2}
}Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy