Sandbox to test various GPU algorithms, image filters, etc. implemented using OpenCL or Cuda
MIT License
Sandbox to test various GPU algorithms, image filters, etc. implemented using OpenCL or Cuda.
Flip X | Flip Y |
---|---|
Blur 3/5/7 | Motion 9 |
---|---|
Sobel | Grayscale |
---|---|
The following code snippet downloads project source code repository, and enters project root folder. Must be executed from the folder where you want to locate project.
$ git clone https://github.com/EgorOrachyov/gpu-sandbox.git
$ cd gpu-sandbox
Attention! On Windows platform building commands must be executed in
x64 Native Tools Command Prompt for VS
.
The following code snippet runs build.py
script, which allows configuring cmake and running of actual build with
selected options. You can specify build directory, build type, number of system threads for build.
Must be executed from project root folder.
$ python ./build.py --build-dir=build --build-type=Release --nt=4
On macOS, you can optionally specify target binaries architecture to build. Pass option --arch
with x86_64
or arm64
respectively. By default, build falls back to CMAKE_SYSTEM_PROCESSOR
specified architecture.
See example bellow, replace <arch>
with desired architecture for your build. Must be executed from project root
folder.
$ python ./build.py --build-dir=build --build-type=Release --nt=4 --arch=<arch>
The following code snippet allows to run executable application of the project to test different algorithms. Must be executed from the directory used for a build in the previous step.
Print help message to get all possible options.
$ ./gpusandbox -h
List all available filters.
$ ./gpusandbox -l
Apply filter to an image and save result.
$ ./gpusandbox --filter=<filter name> --input=<image_src>.png --output=<image_dst>.bmp
@online{spla,
author = {Orachyov, Egor},
title = {Gpu Sandbox: Sandbox to test various GPU algorithms, image filters, etc. implemented using OpenCL or Cuda},
year = 2023,
url = {https://github.com/EgorOrachyov/gpu-sandbox},
note = {Version 1.0.0}
}
Entry | Description |
---|---|
📁 .github |
CI/CD scripts and GitHub related files |
📁 deps |
Third-party project dependencies, stored as submodules |
📁 resource |
Example resource images for testing and playing with |
📁 src |
Private compiled source directory |
📄 CMakeLists.txt |
CMake configuration, add as sub directory to your project |
📄 build.py |
Script to build project sources to executable |
This project licensed under MIT License. License text can be found in the license file.