Marine surface

Visual-Inertial Canoe Dataset

Visual-inertial SLAM in river environments

A 44-minute canoe round trip releasing synchronized stereo imagery, inertial measurements, and GNSS/INS navigation data from a forested Illinois river.

Metadata: Complete Metadata source documented Released 2017

Quick facts

Vehicle type
Marine surface
Environment
Outdoor
Data origin
Real-world
Sensor count
4
Ground truth
Available
Calibration
Parameters Only
Annotations
Not 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 Visual-Inertial Canoe Dataset
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 NovAtel SPAN · SPAN-IGM-A1 1 navigation system 10 Hz Full 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 device specification is 1.2 m RMS horizontal accuracy, but recorded horizontal-position standard deviation is typically 1.5-5.0 m and exceeds 50 m during the subset-6 failure.
Reference notes
The paper treats the released fused position, velocity, and attitude as ground truth and recommends excluding subset 6 from quantitative comparisons.

Calibration and synchronization

Reported calibration level, reproducibility signals, and supporting notes.

Level
Parameters Only
Processed parameters
Available
Raw calibration data
Not reported
Calibration targets
Not reported

Parameters only

Processed calibration parameters are reported, but raw calibration data is not reported.

Notes

Released YAML files provide stereo intrinsics, distortion, stereo extrinsics, rectification parameters, and the camera-to-IMU transformation. A separate calibration archive is downloadable, but its contents were not inspected for this reconciliation.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

GPS position is severely degraded for about five minutes, mostly in subset 6; the authors recommend excluding that subset from comparisons.

Lens flare affects portions of subsets 1, 2, and 5, especially briefly during the turnaround.

Exact automatic exposure duration is unavailable and may differ between the two cameras.

Cameras

Camera 1

Model
IDS UI-3250ML
Setup
Stereo
Count
1
Effective cameras
2
Modality
Rgb
Resolution
800 x 600
Rate
20 Hz
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported
Notes
Two physical cameras use Kowa LM3NC1M 3.5 mm f/2.4 lenses with an approximately 60 cm baseline. Raw Bayer and stereo-rectified RGB images are released; exposure is automatic and its exact duration is unavailable.

IMUs

SPAN integrated IMU

Model
Analog Devices ADIS16488
Count
1
Accel / gyro
200 / 200 Hz

GNSS

NovAtel SPAN-IGM-A1

Model or name
NovAtel SPAN-IGM-A1
Count
1
Rate
10 Hz
GNSS type
GPS_IMU
Position output
Available
Velocity output
Available
Heading output
Available

LiDAR

Not reported

Additional sensors

Not reported

Annotations

Reported annotation availability and task support.

Availability
Not available
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
Miller2018Canoe

BibTeX

@article{Miller2018Canoe, author={Miller, Martin and Chung, Soon-Jo and Hutchinson, Seth}, title={The Visual--Inertial Canoe Dataset}, journal={The International Journal of Robotics Research}, volume={37}, number={1}, pages={13--20}, year={2018}, doi={10.1177/0278364917751842}}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy