Skip to content

Real-Time Decompression and Rasterization of Massive Point Clouds at HPG 2024.

License

Notifications You must be signed in to change notification settings

rahul-goel/pcrhpg24

Repository files navigation

README

Create a build directory to compile the binaries.

mkdir out
cd out
cmake ..
make -j10

To compress a LAS file:

cd out
./preprocess path_to_las.las test.huffman 1

The last argument is 1 if you want to perform Morton Sorting. It can be set to 0 if you want to disable it. We found Morton Sorting to work better in our cases.

The program looks for the file out/test.huffman by default. You can change the path in src/main.cpp accordingly.

To launch the renderer, run the following from the root of the project:

./out/compute_rasterizer
@article{goel2024realtime,
    author={Goel, Rahul and Sch\"{u}tz, Markus and Narayanan, P. J. and Kerbl, Bernhard},
    title={Real-Time Decompression and Rasterization of Massive Point Clouds},
    journal = {Proceedings of the ACM on Computer Graphics and Interactive Techniques},
    year={2024},
    doi = {10.1145/3675373},
    url = {https://rahul-goel.github.io/pcrhpg24/}
}

This work was built on top of Markus' work. Original README is included below for completion.

About

This repository contains the source code for our papers about compute rasterization of point clouds. The project is currently crude and difficult to use, but we plan to add the option to drag&drop your own point clouds after vacation. We will also add a test data set by then.

paper - video

Getting Started

  • Clone the repository
  • Modify ./src/main.cpp so that it loads your own data set.
    • Add a new setting
    • Change Setting setting = settings["..."]; to your own setting.
  • Compile build/ComputeRasterizer.sln with Visual Studio 2022.
  • Run (ctrl + f5)

Currently, only point clouds in LAS format are supported.

Method Location
basic ./modules/compute_loop_las
prefetch ./modules/compute_loop_las2 fastest, each thread fetches 4 points at a time
hqs ./modules/compute_loop_las_hqs High-Quality Shading
LOD ./modules/compute_loop_nodes Support for the Potree LOD format
LOD hqs ./modules/compute_loop_nodes_hqs

About

Real-Time Decompression and Rasterization of Massive Point Clouds at HPG 2024.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •