Upload your data

The parser is currently unavailable, come back in late June or contact us here if you wish to receive a notification email.

We value your feedback. This project is part of ongoing research and any comment regarding the resutls you received is important to us. Please share it by contacting us here.

Drop Here




back to top


What do I need to know?

You can optionally add your data to our publicly available library (research purposes only). The library is freely shared with the community and is merely meant to promote research on this topic. Depending on whether you wish to add your data to the library or not, you'll receive the following results:

  • Don't add data: We will send you 2D visualizations of the results for part of your data.
  • Add data: We will send you all generated files of the rooms and parsed building elements in addition to visualizations, if you help us expand our library. This data will be used only for research purposes and your contribution to the community is appreciated.

Contact us here if you have any questions.


back to top


What kind of data can I upload?

Our method works with the following type of data:

  • We parse point clouds of large-scale indoor spaces. Other type of point clouds (e.g. of buildings under construction or of individual elements) will not produce results.
  • The uploaded file should have the following structure: as many rows as the number of points and 6 columns (X coordinate, Y coordinate, Z coordinate, R color value, G color value, B color value). RGB values should be between 0-255. Click here to download a .txt sample file and here for a .ply sample file.
  • Two file formats are supported: .txt files without headers and .ply files. You can convert most of the formats to any of the supported ones with specialized software. Such examples of open source software in alphabetical order are: Blender, CloudCompare and MeshLab.
  • The maximum file size you can upload is 2GB. Our method does not require very dense point clouds so you can try downsampling them if you exceed the size limit.

If you have any questions or need any help regarding data requirements contact us here.


back to top