Handheld
PennCOSYVIO
Hand-held VIO
Hand-held indoor and outdoor visual-inertial dataset with RGB, fisheye, stereo, IMU, and fiducial-marker ground truth.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- Handheld
- Environment
- Indoor, Outdoor
- Data origin
- Real-world
- Sensor count
- 10
- 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 | Surveyed markers | Surveyed marker network · AprilTag 36h11 | 170 markers | Not reported | Partial · Training Sequences |
Reference provenance
Hybrid
Accuracy and notes
|
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_and_apriltag
Reproducible calibration data
Raw calibration data is reported, so users can reproduce or inspect the calibration process.
Notes
Intrinsic and extrinsic calibration results, calibration imagery, scripts, VI-sensor camera/IMU extrinsics, and Tango device extrinsics are provided.
Known or intentional limitations
Reported constraints and characteristics to check before using the dataset.
Public ground-truth trajectories are provided for two training sequences; two test-sequence trajectories remain hidden for benchmark evaluation.
GoPro cameras use rolling shutters and have relatively large synchronization uncertainty with respect to the IMUs.
Marker-based ground-truth position uncertainty can approach 15 cm near the beginning and end of the route where markers are sparse.
Proprietary Tango Bottom VIO poses are supplied only for training sequences.
Cameras
gopro-ground-truth-cameras
- Model
- GoPro Hero 4 Black
- Setup
- Mono
- Count
- 3
- Effective cameras
- 3
- Modality
- Rgb
- Resolution
- 1920 x 1080
- Rate
- 30 Hz
- Shutter
- Rolling
- Lens type
- Not Reported
- HFOV
- 118.2 deg
- VFOV
- 69.5 deg
tango-bottom-rgb
- Model
- Google Project Tango Yellowstone 7-inch tablet
- Setup
- Mono
- Count
- 1
- Effective cameras
- 1
- Modality
- Rgb
- Resolution
- 1920 x 1080
- Rate
- 30 Hz
- Shutter
- Rolling
- Lens type
- Not Reported
- HFOV
- 52 deg
- VFOV
- 31 deg
tango-top-fisheye
- Model
- Google Project Tango Yellowstone 7-inch tablet
- Setup
- Mono
- Count
- 1
- Effective cameras
- 1
- Modality
- Grayscale
- Resolution
- 640 x 480
- Rate
- 30 Hz
- Shutter
- Global
- Lens type
- Fisheye
- HFOV
- 132 deg
- VFOV
- 100 deg
vi-sensor-stereo
- Model
- Skybotix integrated VI-sensor with Aptina MT9V034 cameras
- Setup
- Stereo
- Count
- 1
- Effective cameras
- 2
- Modality
- Grayscale
- Resolution
- 752 x 480
- Rate
- 20 Hz
- Shutter
- Global
- Lens type
- Not Reported
- HFOV
- 80 deg
- VFOV
- 57 deg
IMUs
Tango Bottom IMU
- Model
- Google Project Tango Yellowstone 7-inch tablet
- Count
- 1
- Accel / gyro
- 128 / 100 Hz
Tango Top IMU
- Model
- Google Project Tango Yellowstone 7-inch tablet
- Count
- 1
- Accel / gyro
- 128 / 100 Hz
Skybotix VI-sensor IMU
- Model
- ADIS16488
- Count
- 1
- Accel / gyro
- 200 / 200 Hz
GNSS
Not reported
LiDAR
Not reported
Additional sensors
Not reported
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.
@inproceedings{Pfrommer2017_penncosyvio,
title = {{{PennCOSYVIO}}: {{A}} Challenging {{Visual Inertial Odometry}} Benchmark},
shorttitle = {{{PennCOSYVIO}}},
booktitle = {2017 {{IEEE International Conference}} on {{Robotics}} and {{Automation}} ({{ICRA}})},
author = {Pfrommer, Bernd and Sanket, Nitin and Daniilidis, Kostas and Cleveland, Jonas},
year = {2017},
month = may,
pages = {3847--3854},
publisher = {{IEEE}},
address = {{Singapore, Singapore}},
doi = {10.1109/ICRA.2017.7989443},
urldate = {2023-10-04},
isbn = {978-1-5090-4633-1}
}Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy