ScanNet++ Dataset

News

October 30, 2025 Released 360° RGB-D panoramic images for 956 scenes (train, val, and semantic test sets). Update the download script to download them.
October 13, 2025 Released the iPhone NVS Benchmark — train on commodity-level captures and test against high-quality DSLR images. Update the download script for the new scenes.
April 30, 2025 Released undistorted DSLR images, 3DGS example codebase, and semantics baselines.
December 20, 2024 ScanNet++ v2 released with 1000+ scenes, more scene types, updated annotations and poses. Check out the Changelog for details.
November 2023 Version 1 of the dataset and NVS and semantic benchmarks released, several updates to the dataset.
October 2023 A ready-to-run dataparser for ScanNet++ is in Nerfstudio now.
September 2023 ScanNet++ website is up! Apply for access to download the data now.

Download the data

To download the data, please create an account, login and create an application. Once your application is approved, you will receive a personalized token to download the data along with further instructions.

Introduction

ScanNet++ is a large scale dataset with 1000+ 3D indoor scenes containing sub-millimeter resolution laser scans, registered 33-megapixel DSLR images, and commodity RGB-D streams from iPhone. The 3D reconstructions are annotated with long-tail and label-ambiguous semantics to benchmark semantic understanding methods, while the coupled DSLR and iPhone captures enable benchmarking of novel view synthesis methods in high-quality and commodity settings.

Benchmarks

Citation

If you use the ScanNet++ data or code please cite:


@inproceedings{yeshwanthliu2023scannetpp,
  title={ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes},
  author={Yeshwanth, Chandan and Liu, Yueh-Cheng and Nie{\ss}ner, Matthias and Dai, Angela},
  booktitle = {Proceedings of the International Conference on Computer Vision ({ICCV})},
  year={2023}
}

License

The ScanNet++ data is released under the ScanNet++ Terms of Use, which you can agree to after signing up.

Privacy

We take privacy very seriously. We have taken great care to ensure that the data is anonymized and does not contain any personally identifiable information. If you notice any privacy concerns, please contact us.