List of Datasets

Handheld, Other

ScanNet++

High-fidelity indoor 3D semantic understanding and novel view synthesis

Original v1 indoor collection of 460 scenes with laser geometry, DSLR imagery and iPhone RGB-D recordings for semantic understanding and view synthesis.

Metadata: Complete Metadata source documented Released 2023

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

Quick facts

Vehicle type
Handheld, Other
Environment
Indoor
Data origin
Real-world
Sensor count
4
Ground truth
Available
Calibration
Parameters Only
Annotations
Available

Sensor overview

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

Cameras

3 reported

IMUs

Not reported

GNSS

Not reported

LiDAR

1 reported

Radar

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 Structure from motion COLMAP + laser geometry alignment Not reported Not reported Partial 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
Reference notes
Image-derived poses gain scale/alignment and photometric refinement from external laser geometry; unreliable iPhone frames are filtered.
Other Other Faro laser-scan scene geometry · Faro Focus Premium 1 laser scanner 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
Aligned laser geometry is also released as a normal point-cloud input. Surface/depth references derive from those same scans; no independent per-frame error guarantee or uniform withheld-test geometry is claimed.

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

Edition-matched v1 scene class lists COLMAP camera files and transforms.json, alongside iPhone pose/intrinsic metadata. Raw calibration sessions and exact DSLR pixel geometry remain unverified.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

This entry covers original v1 (460 scenes); current v2 has over one thousand scenes with revised poses/labels and later panoramic/undistorted products. Current aggregate counts and download sizes do not describe v1.

The reported 17.4 hours concerns iPhone capture only. Scenes are not temporal sequences; DSLR, static laser and iPhone acquisitions are separate sessions rather than one synchronized rigid rig.

Pose alignment reuses imagery and laser geometry. Unreliable iPhone frames are filtered; withheld-test geometry, independent pose/depth accuracy and complete raw calibration remain unestablished.

DSLR dimensions and native iPhone IMU measurements/rates remain unverified. Open-vocabulary/multi-label semantics differ from benchmark subsets; access requires account registration, approved application and a token.

Cameras

Camera 1

Model
Sony Alpha 7 IV
Setup
Fisheye
Count
1
Effective cameras
1
Modality
Rgb
Resolution
Not reported
Rate
Not reported
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported
Notes
33 MP static images with fixed exposure and separate novel-view test captures; exact pixel dimensions are not stated.

Camera 2

Model
Apple iPhone 13 Pro
Setup
Rgbd
Count
1
Effective cameras
2
Modality
Rgbd
Resolution
1920 x 1440
Rate
60 Hz
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported
Notes
RGB dimensions are 1920×1440; auxiliary LiDAR depth is 256×192 at the same stated 60 FPS.

IMUs

Not reported

GNSS

Not reported

LiDAR

Faro Focus Premium

Model or name
Faro Focus Premium
Count
1
Dimensions
3
Rate
Not reported
Channels
Not reported
Range
Not reported

Radar

Not reported

Additional sensors

Not reported

Annotations

Reported annotation availability and task support.

Availability
Available
Format
Open-vocabulary free-text mesh-segment labels, with multi-label and instance annotations
Class count
Not reported
Semantic segmentation Available
Instance segmentation Available
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
yeshwanth2023scannetplusplus

BibTeX

@inproceedings{yeshwanth2023scannetplusplus,
  author = {{Chandan Yeshwanth} and {Yueh-Cheng Liu} and {Matthias Nießner} and {Angela Dai}},
  title = {ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes},
  year = {2023},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
  eprint = {2308.11417},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2308.11417v1}
}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy