Handheld, UGV, Other
VECtor
Event-centric multi-sensor SLAM
Indoor event/stereo/depth/lidar/IMU collection with two recording scales, hardware synchronization and MoCap or FARO-map pose references.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- Handheld, UGV, Other
- Environment
- Indoor
- Data origin
- Real-world
- Sensor count
- 8
- Ground truth
- Available
- Calibration
- Full Raw Calibration Data
- Annotations
- Not reported
Sensor overview
High-level sensor availability before the detailed sensor records below.
Ground truth
Reference scope, method, rate, coverage, and provenance.
| Scope | Method | Reference system | Count | Rate | Coverage |
Reference provenance
|
|---|---|---|---|---|---|---|
| 6DoF | Motion capture | OptiTrack | Not reported | 120 Hz | Partial · Small-Scale Sequences |
Reference provenance
External
|
|
||||||
| 6DoF | Other | FARO survey-grade laser scanner | Not reported | Not reported | Partial · Large-Scale Sequences |
Reference provenance
Hybrid
|
|
||||||
Calibration and synchronization
Reported calibration level, reproducibility signals, and supporting notes.
- Level
- Full Raw Calibration Data
- Processed parameters
- Available
- Raw calibration data
- Available
- Calibration targets
- Static/blinking 9×6 virtual checkerboard, 6×6 Aprilgrid and 7×6 checkerboard; FARO room with checkerboard for camera/lidar calibration.
Reproducible calibration data
Raw calibration data is reported, so users can reproduce or inspect the calibration process.
Notes
Camera intrinsics/extrinsics, camera/IMU, camera/MoCap and camera/lidar parameters plus CAD measurements and raw calibration sequences are released. Kinect uses factory intrinsics/internal depth-color relation; its color data assist calibration only. Five-hour stationary IMU recording supports Allan-variance parameters.
Known or intentional limitations
Reported constraints and characteristics to check before using the dataset.
Depth and lidar streams are mutually split by recording scale.
Regular cameras use 2×2 binning because of bandwidth limits.
Event biases/aperture change with lighting; mean event rates are not fixed camera frequencies.
Kinect color is used for calibration but is not a released recording stream.
Lidar synchronization prose conflicts with the table/timing diagram.
Large-scale truth fits dataset lidar to surveyed FARO maps and lacks a measured trajectory error bound.
Secret submission-benchmark sequences and live service are planned work.
Cameras
Camera 1
- Model
- Prophesee Gen3 CD
- Setup
- Stereo
- Count
- 1
- Effective cameras
- 2
- Modality
- Event
- Resolution
- 640 x 480
- Rate
- Not reported
- Shutter
- Not Reported
- Lens type
- Not Reported
- HFOV
- 67 deg
- VFOV
- 82 deg
- Notes
- One stereo event-camera system (two devices), ~17 cm baseline, Kowa LM5JCM lenses. Asynchronous events; no fixed frame rate. MoCap IR filtering and sequence-dependent bias/aperture settings affect event statistics.
Camera 2
- Model
- FLIR Grasshopper3
- Setup
- Stereo
- Count
- 1
- Effective cameras
- 2
- Modality
- Grayscale
- Resolution
- 1224 x 1024
- Rate
- 30 Hz
- Shutter
- Global
- Lens type
- Not Reported
- HFOV
- 61 deg
- VFOV
- 82 deg
- Notes
- One stereo system (two cameras), ~17 cm baseline, Kowa LM6JC lenses; 2×2 binning due to bandwidth. This regular stereo pair is physically separate from the event pair.
Camera 3
- Model
- Azure Kinect
- Setup
- Mono
- Count
- 1
- Effective cameras
- 1
- Modality
- Depth
- Resolution
- 640 x 576
- Rate
- 30 Hz
- Shutter
- Not Reported
- Lens type
- Not Reported
- HFOV
- 75 deg
- VFOV
- 65 deg
- Notes
- Only depth is recorded, narrow unbinned mode, up to 3.86 m. Present in small-scale sequences, omitted from large-scale sequences. Color is used only for calibration; its rolling shutter is not assigned to the depth channel.
IMUs
Xsens AHRS inertial stream
- Model
- XSens MTi-30 AHRS
- Count
- 1
- Accel / gyro
- 200 / 200 Hz
GNSS
Not reported
LiDAR
Ouster OS0-128
- Model or name
- Ouster OS0-128
- Count
- 1
- Dimensions
- 3
- Rate
- 10 Hz
- Channels
- 128
- Range
- 50 m
Radar
Not reported
Additional sensors
Magnetometer
- Name
- Xsens AHRS magnetic-field output
- Count
- 1
- Notes
- Three-axis magnetometer belongs to the recorded XSens MTi-30 AHRS device; no separate physical device count or component model is inferred, and magnetic-channel cadence is not separately specified.
Annotations
Reported annotation availability and task support.
- Availability
- Not reported
- Format
- Not reported
- Class count
- Not reported
Semantic segmentation
Not reported
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.
@misc{gao2022vector,
title={VECtor: A Versatile Event-Centric Benchmark for Multi-Sensor SLAM},
author={Gao, Ling and Liang, Yuxuan and Yang, Jiaqi and Wu, Shaoxun and Wang, Chenyu and Chen, Jiaben and Kneip, Laurent},
year={2022},
eprint={2207.01404},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2207.01404v1}
}Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy