💦 OpenSplat

A free and open source implementation of 3D gaussian splatting written in C++, focused on being portable, lean and fast.

OpenSplat takes camera poses + sparse points in COLMAP, OpenSfM, ODM or nerfstudio project format and computes a scene file (.ply or .splat) that can be later imported for viewing, editing and rendering in other software.

Graphics card recommended, but not required! OpenSplat runs the fastest on NVIDIA, AMD and Apple (Metal) GPUs, but can also run entirely on the CPU (~100x slower).

Commercial use allowed and encouraged under the terms of the AGPLv3. ✅

We even have a song 🎵

Getting Started

If you're on Windows, you can buy the pre-built program. This saves you time and helps support the project ❤️. Then jump directly to the run section. As an alternative, check the build section below.

If you're on macOS or Linux check the build section below.

Build

You can build OpenSplat with or without GPU support.

Requirements for all builds:

OpenCV: sudo apt install libopencv-dev should do it.
libtorch: See instructions below.

CPU

For libtorch visit https://pytorch.org/get-started/locally/ and select your OS, for package select "LibTorch". For compute platform you can select "CPU".

Then:

git clone https://github.com/pierotofy/OpenSplat OpenSplat
cd OpenSplat
mkdir build && cd build
cmake -DCMAKE_PREFIX_PATH=/path/to/libtorch/ .. && make -j$(nproc)

CUDA

Additional requirement:

CUDA: Make sure you have the CUDA compiler (nvcc) in your PATH and that nvidia-smi is working. https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html

For libtorch visit https://pytorch.org/get-started/locally/ and select your OS, for package select "LibTorch". Make sure to match your version of CUDA if you want to leverage GPU support in libtorch.

Then:

git clone https://github.com/pierotofy/OpenSplat OpenSplat
cd OpenSplat
mkdir build && cd build
cmake -DCMAKE_PREFIX_PATH=/path/to/libtorch/ .. && make -j$(nproc)

ROCm via HIP

Additional requirement:

ROCm: Make sure you have the ROCm installed at /opt/rocm. https://rocm.docs.amd.com/projects/install-on-linux/en/latest/tutorial/quick-start.html

For libtorch visit https://pytorch.org/get-started/locally/ and select your OS, for package select "LibTorch". Make sure to match your version of ROCm (5.7) if you want to leverage AMD GPU support in libtorch.

Then:

git clone https://github.com/pierotofy/OpenSplat OpenSplat
cd OpenSplat
mkdir build && cd build
export PYTORCH_ROCM_ARCH=gfx906
cmake -DCMAKE_PREFIX_PATH=/path/to/libtorch/ -DGPU_RUNTIME="HIP" -DHIP_ROOT_DIR=/opt/rocm -DOPENSPLAT_BUILD_SIMPLE_TRAINER=ON ..
make

In addition, you can leverage Jinja to build the project

cmake -GNinja -DCMAKE_PREFIX_PATH=/path/to/libtorch/ -DGPU_RUNTIME="HIP" -DHIP_ROOT_DIR=/opt/rocm -DOPENSPLAT_BUILD_SIMPLE_TRAINER=ON ..
jinja

Windows

There's several ways to build on Windows, but this particular configuration has been confirmed to work:

Then run:

"C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Auxiliary/Build/vcvars64.bat"
git clone https://github.com/pierotofy/OpenSplat OpenSplat
cd OpenSplat
md build
cd build
cmake -DCMAKE_PREFIX_PATH=C:/path_to/libtorch_2.1.2_cu11.8/libtorch -DOPENCV_DIR=C:/path_to/OpenCV_4.9.0/build -DCMAKE_BUILD_TYPE=Release ..
cmake --build . --config Release

Optional: Edit cuda target (only if required) before cmake --build .

C:/path_to/OpenSplat/build/gsplat.vcxproj for example: arch=compute_75,code=sm_75

macOS

If you're using Homebrew, you can install Cmake/OpenCV/Pytorch by running:

brew install cmake
brew install opencv
brew install pytorch

You will also need to install Xcode and the Xcode command line tools to compile with metal support (otherwise, OpenSplat will build with CPU acceleration only):

Install Xcode from the Apple App Store.
Install the command line tools with xcode-select --install. This might do nothing on your machine.
If xcode-select --print-path prints /Library/Developer/CommandLineTools,then run sudo xcode-select --switch /Applications/Xcode.app/Contents/Developer.

Then run:

git clone https://github.com/pierotofy/OpenSplat OpenSplat
cd OpenSplat
mkdir build && cd build
cmake -DCMAKE_PREFIX_PATH=/path/to/libtorch/ -DGPU_RUNTIME=MPS .. && make -j$(sysctl -n hw.logicalcpu)
./opensplat

If building CPU-only, remove -DGPU_RUNTIME=MPS.

⚠️ You will probably get a libc10.dylib can’t be opened because Apple cannot check it for malicious software error on first run. Open System Settings and go to Privacy & Security and find the Allow button. You might need to repeat this several times until all torch libraries are loaded.

Docker Build

CUDA

Navigate to the root directory of OpenSplat repo that has Dockerfile and run the following command to build the Docker image:

docker build -t opensplat .

The -t flag and other --build-arg let you tag and further customize your image across different ubuntu versions, CUDA/libtorch stacks, and hardware accelerators. For example, to build an image with Ubuntu 22.04, CUDA 12.1.1, libtorch 2.2.1, and support for CUDA architectures 7.0 and 7.5, run the following command:

docker build \
  -t opensplat:ubuntu-22.04-cuda-12.1.1-torch-2.2.1 \
  --build-arg UBUNTU_VERSION=22.04 \
  --build-arg CUDA_VERSION=12.1.1 \
  --build-arg TORCH_VERSION=2.2.1 \
  --build-arg CMAKE_CUDA_ARCHITECTURES="70;75;80" \
  --build-arg CMAKE_BUILD_TYPE=Release .

ROCm via HIP

Navigate to the root directory of OpenSplat repo that has Dockerfile and run the following command to build the Docker image:

docker build \
  -t opensplat \
  -f Dockerfile.rocm .

The -t flag and other --build-arg let you tag and further customize your image across different ubuntu versions, CUDA/libtorch stacks, and hardware accelerators. For example, to build an image with Ubuntu 22.04, CUDA 12.1.1, libtorch 2.2.1, ROCm 5.7.1, and support for ROCm architectures gfx906, run the following command:

docker build \
  -t opensplat:ubuntu-22.04-cuda-12.1.1-libtorch-2.2.1-rocm-5.7.1-llvm-16 \
  --build-arg UBUNTU_VERSION=22.04 \
  --build-arg CUDA_VERSION=12.1.1 \
  --build-arg TORCH_VERSION=2.2.1 \
  --build-arg ROCM_VERSION=5.7.1 \
  --build-arg PYTORCH_ROCM_ARCH="gfx906" \
  --build-arg CMAKE_BUILD_TYPE=Release .

Note: If you want to use ROCm 6.x, you need to switch to AMD version of pytorch docker as a base layer to build:

docker build \
  -t opensplat:ubuntu-22.04-libtorch-2.1.2-rocm-6.0.2 \
  -f Dockerfile.rocm6 .

Run

To get started, download a dataset and extract it to a folder: [ banana ] [ train ] [ truck ]

Then run from a command line prompt:

Windows

cd c:\path\to\opensplat
opensplat.exe c:\path\to\banana -n 2000

macOS / Linux

cd build
./opensplat /path/to/banana -n 2000

The program will generate an output splat.ply file which can then be dragged and dropped in one of the many viewers such as https://playcanvas.com/viewer. You can also edit/cleanup the scene using https://playcanvas.com/supersplat/editor. The program will also output a cameras.json file in the same directory which can be used by some viewers.

To run on your own data, choose the path to an existing COLMAP, OpenSfM, ODM or nerfstudio project. The project must have sparse points included (random initialization is not supported, see #7).

There's several parameters you can tune. To view the full list:

./opensplat --help

Compression

To generate compressed splats (.splat files), use the -o option:

./opensplat /path/to/banana -o banana.splat

AMD GPU Notes

To train a model with AMD GPU using docker container, you can use the following command as a reference:

Launch the docker container with the following command:

docker run -it -v ~/data:/data --device=/dev/kfd --device=/dev/dri opensplat:ubuntu-22.04-libtorch-2.1.2-rocm-6.0.2 bash

Inside the docker container, run the following command to train the model:

export HIP_VISIBLE_DEVICES=0
export HSA_OVERRIDE_GFX_VERSION=10.3.0  # AMD RX 6700 XT workaround 
cd /code/build
./opensplat /data/banana -n 2000

Project Goals

We recently released OpenSplat, so there's lots of work to do.

Support for running on AMD cards (more testing needed)
Improve speed / reduce memory usage
Distributed computation using multiple machines
Real-time training viewer output
Automatic filtering
Your ideas?

https://github.com/pierotofy/OpenSplat/issues?q=is%3Aopen+is%3Aissue+label%3Aenhancement

Contributing

We welcome contributions! Pull requests are welcome.

GPU Memory Notes

A single gaussian takes ~2000 bytes of memory, so currenly you need ~2GB of GPU memory for each million gaussians.

Credits

The methods used in OpenSplat are originally based on splatfacto.

License

The code in this repository, unless otherwise noted, is licensed under the AGPLv3.

Name		Name	Last commit message	Last commit date
Latest commit History 313 Commits
.github/workflows		.github/workflows
rasterizer		rasterizer
.gitignore		.gitignore
CMakeLists.txt		CMakeLists.txt
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
Dockerfile		Dockerfile
Dockerfile.rocm		Dockerfile.rocm
Dockerfile.rocm6		Dockerfile.rocm6
LICENSE.txt		LICENSE.txt
README.md		README.md
VERSION		VERSION
colmap.cpp		colmap.cpp
colmap.hpp		colmap.hpp
constants.hpp		constants.hpp
cv_utils.cpp		cv_utils.cpp
cv_utils.hpp		cv_utils.hpp
gsplat.hpp		gsplat.hpp
input_data.cpp		input_data.cpp
input_data.hpp		input_data.hpp
kdtree_tensor.cpp		kdtree_tensor.cpp
kdtree_tensor.hpp		kdtree_tensor.hpp
model.cpp		model.cpp
model.hpp		model.hpp
nerfstudio.cpp		nerfstudio.cpp
nerfstudio.hpp		nerfstudio.hpp
opensfm.cpp		opensfm.cpp
opensfm.hpp		opensfm.hpp
opensplat.cpp		opensplat.cpp
opensplat.hpp		opensplat.hpp
optim_scheduler.cpp		optim_scheduler.cpp
optim_scheduler.hpp		optim_scheduler.hpp
point_io.cpp		point_io.cpp
point_io.hpp		point_io.hpp
project_gaussians.cpp		project_gaussians.cpp
project_gaussians.hpp		project_gaussians.hpp
rasterize_gaussians.cpp		rasterize_gaussians.cpp
rasterize_gaussians.hpp		rasterize_gaussians.hpp
simple_trainer.cpp		simple_trainer.cpp
spherical_harmonics.cpp		spherical_harmonics.cpp
spherical_harmonics.hpp		spherical_harmonics.hpp
ssim.cpp		ssim.cpp
ssim.hpp		ssim.hpp
tensor_math.cpp		tensor_math.cpp
tensor_math.hpp		tensor_math.hpp
tile_bounds.hpp		tile_bounds.hpp
utils.cpp		utils.cpp
utils.hpp		utils.hpp

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

💦 OpenSplat

Getting Started

Build

CPU

CUDA

ROCm via HIP

Windows

macOS

Docker Build

CUDA

ROCm via HIP

Run

Windows

macOS / Linux

Compression

AMD GPU Notes

Project Goals

Contributing

GPU Memory Notes

Credits

License

About

Releases 9

Packages

Contributors 12

Languages

License

pierotofy/OpenSplat

Folders and files

Latest commit

History

Repository files navigation

💦 OpenSplat

Getting Started

Build

CPU

CUDA

ROCm via HIP

Windows

macOS

Docker Build

CUDA

ROCm via HIP

Run

Windows

macOS / Linux

Compression

AMD GPU Notes

Project Goals

Contributing

GPU Memory Notes

Credits

License

About

Topics

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases 9

Packages 0

Contributors 12

Languages

Packages