xelatihy / yocto-gl

Yocto/GL: Tiny C++ Libraries for Data-Driven Physically-based Graphics
https://xelatihy.github.io/yocto-gl
2.83k stars 207 forks source link
3d-graphics computer-graphics cplusplus cpp gltf graphics path-tracer path-tracing physically-based-rendering rendering

Yocto/GL: Tiny C++ Libraries for Data-Oriented Physically-based Graphics

Yocto/GL is a collection of small C++17 libraries for building physically-based graphics algorithms released under the MIT license. Yocto/GL is written in a deliberately data-oriented style for ease of development and use. Yocto/GL is split into small libraries to make code navigation easier. See each header file for documentation.

You can see Yocto/GL in action in the following applications written to test the library:

Here are some test images rendered with the path tracer. More images are included in the project site.

Example materials: matte, plastic, metal, glass, subsurface, normal mapping

Example shapes: procedural shapes, Catmull-Clark subdivision, hairs, displacement mapping

Image rendered with Yocto/GL path tracer. Model by Disney Animation Studios.

Design Considerations

Yocto/GL follows a "data-oriented programming model" that makes data explicit. Data is stored in simple structs and accessed with free functions or directly. All data is public, so we make no attempt at encapsulation. We do this since this makes Yocto/GL easier to extend and quicker to learn, with a more explicit data flow that is easier when writing parallel code. Since Yocto/GL is mainly used for research and teaching, explicit data is both more hackable and easier to understand.

Nearly all objects in Yocto/GL have value semantic. This means that everything can be trivially copied and serialized and there is no need for memory management. While this has the drawback of potentially introducing spurious copies, it does have the benefit of ensuring that no memory corruption can occur, which turned out was a major issue for novice C++ users, even in a very small library like this one.

In terms of code style we prefer a functional approach rather than an object oriented one, favoring free functions to class methods. All functions and data are defined in the yocto namespace so libraries can call each others trivially.

The use of templates in Yocto was the reason for many refactoring, going from no template to heavy template use. At this point, Yocto uses some templates for readability. In the future, we will increase the use of templates in math code, while keeping many APIs explicitly typed.

For error handling in IO we either return status object or an interface that uses boolean flags and error strings. Internally exceptions are used when used by external libraries, but otherwise no exception are used. At the moment, exceptions are only used to report "programmer errors", namely when preconditions or post conditions are violated in functions, just lime the standard library does.

License

The library is released under the MIT license. We include various external dependencies in the distribution that each have thir own license, compatible with the chosen one.

Compilation

This library requires a C++17 compiler and is know to compiled on OsX (Xcode >= 11), Windows (MSVC >= 2019) and Linux (gcc >= 9, clang >= 9).

You can build the example applications using CMake with mkdir build; cd build; cmake ..; cmake --build .

Yocto/GL required dependencies are included in the distribution and do not need to be installed separately.

Yocto/GL optionally supports building OpenGL demos. OpenGL support is enabled by defining the cmake option YOCTO_OPENGL. OpenGL dependencies are included in this repo.

Yocto/GL optionally supports the use of Intel's Embree for ray casting. See the main CMake file for how to link to it. Embree support is enabled by defining the cmake option YOCTO_EMBREE. Embree needs to be installed separately.

Yocto/GL optionally supports the use of Intel's Open Image Denoise for denoising. See the main CMake file for how to link to it. Open Image Denoise support is enabled by defining the cmake option YOCTO_DENOISE. OIDN needs to be installed separately.