LidarScout

Direct Out-of-Core Rendering of Massive Point Clouds

This is the official implementation of LidarScout (High-Performance Graphics 2025). This is the repository for training and evaluating the model. For the viewer and an executable, see https://github.com/cg-tuwien/lidarscout

Setup

We tested this repository on Windows 10/11 with CUDA 12.4. We recommend installing the dependencies with venv and Pip.

# clone this repo
git clone git@github.com:cg-tuwien/lidarscout_training.git

# go into the cloned dir
cd lidarscout_training

1. Create the virtual environment with: `python -m venv .venv`
2. Activate the virtual environment in a terminal with a matching script in `.venv\Scripts\`
   1. Windows Powershell: [Enable script execution via PS with admin rights](https://learn.microsoft.com/en-us/powershell/module/microsoft.powershell.core/about/about_execution_policies?view=powershell-7.5): `Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser`. Then activate the virtual environment in a normal terminal with `.venv\Scripts\Activate.ps1`.
   2. Windows CMD: Run `.venv\Scripts\activate.bat`
   3. Linux: Run `.venv\Scripts\activate`
3. Install all necessary dependencies with: `pip install -r requirements.txt`

Datasets, Model and Results

Datasets for training (extract to have the bins folder and several text files in [repo_root]/datasets/laz_minimal/): https://users.cg.tuwien.ac.at/perler/lidarscout/laz_minimal.zip

Models for evaluation (extract to have e.g. ipes_cnn in [repo_root]/models): https://users.cg.tuwien.ac.at/perler/lidarscout/models.zip

Results (extract to [repo_root]/results):

• (main) https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_cnn_rgb.zip
• (main without color) https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_cnn.zip
• (best with extra training data) https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_cnn_allstar.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_cnn_colorizer.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_cnn_only_lin.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_cnn_only_nn.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_interp_cubic.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_interp_linear.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_interp_nearest.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_rast.zip
• https://users.cg.tuwien.ac.at/perler/lidarscout/results/ipes_unet.zip

Create your own datasets

1. WARNING: These tools are not exactly clean and robust. Open an issue if you run into problems.
2. Find a nice scan in LAS or LAZ format. There are public repositories like Opentopography.
3. Follow the instructions in [repo_root]/datasets/tools/00_readme.md
4. Create new dataset splits or add yours to the textfiles in [repo_root]/datasets/laz_minimal/.

Run the Training, Reconstruction and Evaluation

Make sure, you have activated the virtual environment as described above. Now, you can run the training, reconstruction and evaluation simply with

python ipes.py

Training takes about 30-45 minutes on a NVIDIA RTX 3090. The first epoch will take longer since we need to create the image cache from the point sub-samples.

Logging during training with Tensorboard is enabled by default. You can start a Tensorboard server with:

tensorboard --logdir models

To run other model variants for ablation and comparison, modify the parameters in the YAML configs in [repo_root]/configs or override then in ipes.py and ipes_interp.py. Run those again for a train, reconstruct, evaluate cycle:

python ipes.py         # learned
python ipes_interp.py  # un-learned, skips training

You can gather all the evaluation statistics from the XLSX files for easy copy&paste by running

python gather_table_data.py

At the end of the training, the model will be traced, creating a TorchScript model. This file is stored in [repo_root]/models as .pt. You can transfer this to the LidarScout viewer. See the heightmap_interp sub-project for the usage in LibTorch.

Trouble Shooting

Pip might fail when creating the environment. If so, try installing the Pip packages from the pps.yml manually.

On Windows, Pip install may raise a "Microsoft Visual C++ 14.0 or greater is required. Get it with "Microsoft C++ Build Tools" error. In this case, install the MS Visual Studio build tools, as described on Stackoverflow.

Citation

If you use our work, please cite our paper:

@inproceedings{erler2025lidarscout,
  booktitle = {High-Performance Graphics - Symposium Papers},
  editor = {Knoll, Aaron and Peters, Christoph},
  title = {{LidarScout: Direct Out-of-Core Rendering of Massive Point Clouds}},
  author = {Erler, Philipp and Herzberger, Lukas and Wimmer, Michael and Schütz, Markus},
  year = {2025},
  publisher = {The Eurographics Association},
  ISSN = {2079-8687},
  ISBN = {978-3-03868-291-2},
  DOI = {10.2312/hpg.20251170}
}