Lightweight Multidimensional Adaptive Sampling for GPU Ray Tracing

Sources codes of the Lightweight Multidimensional Adaptive Sampling for GPU Ray Tracing project. We extended Optix samples to support the proposed parallel multidimensional sampling and reconstruction. In particular, we added five new samples: optixMotionBlur, optixDepthOfField, optixAmbientOcclusion, optixPathTracer, and optixDirectLighting.

Compilation

We compiled the project with Visual Studio 2019 (x64), but it should work also with other compilers using CMake.

Usage

There are three sample scenes in SDK/data: pool, cornell-box, and chess. We use env file format for the configuration. Besides scene configuration, we can also configure sampling:

Sampler {
    mdas true # use mdas or qmc
    samples 8 # number of saples
}

Mdas { # mdas parameters (see paper for details)
    scaleFactor 1 
    alpha 0.25
    bitsPerDim 1
    extraImgBits 8 
}

We simply use the env file as argument to run the sample:

./optixMotionBlur.exe ../../../data/pool/pool.env
./optixDepthOfField.exe ../../../data/chess/chess.env
./optixPathTracer.exe ../../../data/cornell-box/cornell-box.env

There test scripts in SDK/data/Scripts that we used to generate the paper results.

License

The additional code is released into the public domain.

Citation

If you use this code, please cite the paper:

@Article{Meister2022,
  author = {Daniel Meister and Toshiya Hachisuka},
  title = {{Lightweight Multidimensional Adaptive Sampling for GPU Ray Tracing}},
  journal = {Journal of Computer Graphics Techniques (JCGT)},
  volume = {11},
  number = {3},
  pages = {46--64},
  year = {2022},
}