UAV

Zurich Urban MAV

MAV VIO

Outdoor quadcopter visual-inertial dataset with RGB imagery and visual-pose ground truth.

Metadata: Complete Metadata source documented Released 2017

Quick facts

Vehicle type
UAV
Environment
Outdoor
Data origin
Real-world
Sensor count
5
Ground truth
Available
Calibration
Full Raw Calibration Data
Annotations
Available

Sensor overview

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

Ground truth

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

Ground-truth references for Zurich Urban MAV
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.
3D position Photogrammetry Not reported Not reported 1 Hz 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.
Hybrid
Accuracy and notes
Reference notes
Metric camera positions from Pix4D reconstruction using loop closures, GPS initialization, manual control points, and bundle adjustment. Separate visually verified aerial-to-Street-View matches support topological localization.

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

Thirty checkerboard images and OpenCV pinhole intrinsics are included; the camera-to-body translation is reported in the paper.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

Photogrammetric ground truth provides camera positions rather than an independently measured external 6DoF reference trajectory.

Urban-canyon GPS shadowing and multipath produce reported RMS errors of 2.22 m, 3.76 m, and 5.46 m along the three position axes relative to the reconstructed path.

The image-to-onboard-data temporal offset was set once at the beginning of recording rather than estimated independently for every frame.

The GoPro Hero 4 rolling shutter has a 30 ms top-to-bottom readout time.

Cameras

mono-rgb

Model
GoPro Hero 4
Setup
Mono
Count
1
Effective cameras
1
Modality
Rgb
Resolution
1920 x 1080
Rate
30 Hz
Shutter
Rolling
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported

IMUs

Pixhawk PX4 autopilot IMU

Model
Not reported
Count
1
Accel / gyro
10 / 10 Hz

GNSS

Onboard GPS receiver

Model or name
Onboard GPS receiver
Count
1
Rate
Not reported
GNSS type
GPS
Position output
Available
Velocity output
Not available
Heading output
Not reported

LiDAR

Not reported

Additional sensors

Barometer

Name
PX4 autopilot barometer
Count
1

Odometry

Name
Pixhawk PX4 autopilot pose estimate
Count
1

Annotations

Reported annotation availability and task support.

Availability
Available
Format
GroundTruthAGM.csv aerial-to-Street-View image correspondences
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
Majdik2017

BibTeX

@article{Majdik2017,
  title = {The {{Zurich}} Urban Micro Aerial Vehicle Dataset},
  author = {Majdik, Andr{\'a}s L and Till, Charles and Scaramuzza, Davide},
  year = {2017},
  month = mar,
  journal = {The International Journal of Robotics Research},
  volume = {36},
  number = {3},
  pages = {269--273},
  issn = {0278-3649, 1741-3176},
  doi = {10.1177/0278364917702237},
  urldate = {2023-10-04},
  langid = {english}
}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy