List of Datasets

Automotive

Oxford RobotCar

Self-driving car

Long-term automotive dataset with RGB cameras, trinocular camera setup, SICK laser sensors, and GNSS/IMU ground truth.

Metadata: Complete Metadata source documented Released 2017

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

Quick facts

Vehicle type
Automotive
Environment
Outdoor
Data origin
Real-world
Sensor count
12
Ground truth
Available
Calibration
Full Raw Calibration Data
Annotations
Not reported

Sensor overview

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

Cameras

6 reported

IMUs

1 reported

GNSS

1 reported

LiDAR

3 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 Not reported Not reported Not reported Partial · 72 Traversals 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.
Hybrid
Accuracy and notes
Accuracy notes
The maintained release describes the post-processed RTK reference as centimetre-accurate.
Reference notes
Globally consistent post-processed RTK GNSS/INS reference poses are provided for 72 traversals. Original SPAN-CPT GPS/INS and released stereo VO are useful pose sources but are not treated as uniformly reliable ground truth.

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
checkerboard

Reproducible calibration data

Raw calibration data is reported, so users can reproduce or inspect the calibration process.

Notes

Camera intrinsics, rectification and undistortion lookup tables, and precise sensor extrinsics are provided. Original Grasshopper2 checkerboard images support recalibration.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

RTK ground truth covers 72 traversals and excludes the six RobotCar Seasons runs; it is not available for every recording.

The original GPS/INS solution varies with satellite reception and can drift or develop large positioning errors; its quality and status fields must be checked.

Released stereo visual odometry is locally smooth but drifts at large scale and is affected by scene content and exposure.

Provided long-term extrinsic calibrations are best estimates and are not guaranteed to be exact for every traversal.

Distributed LiDAR points are not motion-compensated, and Bumblebee software auto-exposure occasionally produces overexposed frames.

Cameras

grasshopper2-surround

Model
Point Grey Grasshopper2 GS2-FW-14S5C-C
Setup
Mono
Count
3
Effective cameras
3
Modality
Rgb
Resolution
1024 x 1024
Rate
11.1 Hz
Shutter
Global
Lens type
Fisheye
HFOV
180 deg
VFOV
Not reported

bumblebee-xb3-trinocular

Model
Point Grey Bumblebee XB3 BBX3-13S2C-38
Setup
Trinocular
Count
1
Effective cameras
3
Modality
Rgb
Resolution
1280 x 960
Rate
16 Hz
Shutter
Global
Lens type
Not Reported
HFOV
66 deg
VFOV
Not reported

IMUs

NovAtel SPAN-CPT ALIGN

Model
NovAtel SPAN-CPT ALIGN
Count
1
Accel / gyro
50 / 50 Hz

GNSS

NovAtel SPAN-CPT ALIGN

Model or name
NovAtel SPAN-CPT ALIGN
Count
1
Rate
5 Hz
GNSS type
GPS_GLONASS
Position output
Available
Velocity output
Not available
Heading output
Not reported

LiDAR

SICK LMS-151

Model or name
SICK LMS-151
Count
2
Dimensions
2
Rate
50 Hz
Channels
Not reported
Range
50 m

SICK LD-MRS

Model or name
SICK LD-MRS
Count
1
Dimensions
3
Rate
12.5 Hz
Channels
4
Range
50 m

Additional sensors

Odometry

Name
Bumblebee XB3 wide-baseline stereo visual odometry
Count
1

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.

Citation key
Maddern2017_1year1000km

BibTeX

@article{Maddern2017_1year1000km,
  title = {1 Year, 1000 Km: {{The Oxford RobotCar}} Dataset},
  shorttitle = {1 Year, 1000 Km},
  author = {Maddern, Will and Pascoe, Geoffrey and Linegar, Chris and Newman, Paul},
  year = {2017},
  month = jan,
  journal = {The International Journal of Robotics Research},
  volume = {36},
  number = {1},
  pages = {3--15},
  issn = {0278-3649, 1741-3176},
  doi = {10.1177/0278364916679498},
  urldate = {2023-10-04},
  langid = {english}
}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy