ScanNetV2

The dataset used in the paper is ScanNetV2, a real-world dataset for indoor scenes. It contains 1205 training scenes and 312 testing scenes, with instance-level object bounding box labels provided for 18 categories.

Data and Resources

Cite this as

Jiahao Lu, Jiacheng Deng, Tianzhu Zhang (2024). Dataset: ScanNetV2. https://doi.org/10.57702/z9chl0jz

DOI retrieved: December 2, 2024

Additional Info

Field Value
Created December 2, 2024
Last update December 2, 2024
Defined In https://doi.org/10.48550/arXiv.2004.01170
Citation
  • https://doi.org/10.48550/arXiv.2107.14230
  • https://doi.org/10.1109/ICME52920.2022.9859996
  • https://doi.org/10.48550/arXiv.1909.13603
  • https://doi.org/10.48550/arXiv.2211.09375
Author Jiahao Lu
More Authors
Jiacheng Deng
Tianzhu Zhang
Homepage https://www.scannet.org/