List of Datasets

Other

VinT-6D

6D object-in-hand pose estimation

Synthetic and real visual, tactile and proprioceptive data for multi-finger robotic object-in-hand pose estimation.

Metadata: Partial Metadata needs review

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

Download

Not reported

Code

Not reported

Supplementary

Not reported

Quick facts

Vehicle type
Other
Environment
Not reported
Data origin
Hybrid
Sensor count
3
Ground truth
Available
Calibration
Not reported
Annotations
Available

Sensor overview

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

Cameras

3 reported

IMUs

Not reported

GNSS

Not reported

LiDAR

Not reported

Radar

Not reported

Pitot tube

Not 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 Motion capture Vicon Not reported Not reported 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.
External
Accuracy notes
Authors claim submillimeter hand-pose accuracy; no numerical object-pose error certificate is transferred from that claim.
Reference notes
VinT-Real object poses use markers on custom rigid fixtures; fixture models are promised publicly. Reference apparatus is excluded from the input-sensor inventory.
Not reported Motion capture Vicon Not reported Not reported 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.
External
Accuracy notes
Submillimeter hand-pose accuracy is claimed without a scalar position/orientation error specification.
Reference notes
VinT-Real hand-pose reference uses a marker-equipped flange on the Trx hand.
6DoF Other Not reported Not reported Not reported 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.
Derived
Reference notes
VinT-Sim records simulated object, hand and finger-knuckle poses during grasp generation. This reference is simulator state, not a second physical motion capture system.

Calibration and synchronization

Reported calibration level, reproducibility signals, and supporting notes.

Level
Not reported
Processed parameters
Not reported
Raw calibration data
Not reported
Calibration targets
Not reported

Calibration data

Calibration reproducibility details are not reported.

Notes

Chessboard intrinsic/extrinsic camera calibration, distortion correction and depth reprojection are described; availability of calibration files or raw calibration sequences is not established by the paper.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

Limited object and scene diversity acknowledged by the authors.

Finger-induced occlusions range from 10 to 90 percent in the synthetic split.

Cameras

real-stereo

Model
Not reported
Setup
Stereo
Count
1
Effective cameras
2
Modality
Rgb
Resolution
Not reported
Rate
Not reported
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported
Notes
Real split left/right industrial color camera images; right-camera capture is reported at 1200 x 960 and aligned RGB-depth products at 640 x 400. These do not establish a common stereo release resolution.

real-depth

Model
Azure Kinect
Setup
Mono
Count
1
Effective cameras
1
Modality
Depth
Resolution
Not reported
Rate
Not reported
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported
Notes
Real split includes depth and auxiliary IR from this one TOF device. The paper reports native depth 640 x 565 and aligned images 640 x 400; the reported native dimensions are retained without correction.

IMUs

Not reported

GNSS

Not reported

LiDAR

Not reported

Radar

Not reported

Pitot tube

Not reported

Additional sensors

Other sensor

Name
Trx-hand tactile arrays
Count
1
Notes
Real split touch readings from curved piezoresistive arrays across fingertips, pulp and palm; 620 taxels per Trx hand. Taxels are sensing elements, not 620 physical device groups. The reported response exceeding 250 Hz is a capability, not an established release cadence.

Other sensor

Name
Trx-hand proprioception
Count
1
Notes
Finger joint angles are listed as dataset inputs; the full apparatus has two ABB arms and Trx hands, which does not establish two released joint streams.

Annotations

Reported annotation availability and task support.

Availability
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
wan2024vint6d

BibTeX

@inproceedings{wan2024vint6d,
  title = {VinT-6D: A Large-Scale Object-in-hand Dataset from Vision, Touch and Proprioception},
  author = {Wan, Zhaoliang and Ling, Yonggen and Yi, Senlin and Qi, Lu and Lee, Wangwei and Lu, Minglei and Yang, Sicheng and Teng, Xiao and Lu, Peng and Yang, Xu and Yang, Ming-Hsuan and Cheng, Hui},
  booktitle = {Proceedings of the 41st International Conference on Machine Learning},
  series = {Proceedings of Machine Learning Research},
  volume = {235},
  year = {2024}
}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy