Skip to content
This repository has been archived by the owner on Oct 11, 2023. It is now read-only.
/ fastwfcnet Public archive

WaveFunctionCollapse in C# .Net

License

Notifications You must be signed in to change notification settings

ShyRed/fastwfcnet

Repository files navigation

CI Download
Build status Build Status NuGet

FastWfcNet is a library that generates 2D output data that is locally similar to the 2D input data. This implementation is a port of the fast-wcf library to C# .NET. Several modifications have been made to use C# / .NET features where possible.

Highlights

  • .NET Standard 2.0 library with no dependencies
  • Simple WinForms Demo Application (visit mxgmn's repository for example images / tilemaps)

Features

  • 2D Overlapping model with improved floor pattern detection
  • 2D Tiling model with support for tiles that have no symmetry
  • Fully documented codebase

The Demo Application

The demo application allows you to try out the algorithm and get a feeling for what the different options do. The application's source code deals with handling bitmap images and as such can be used as a reference point when using the library to work with bitmap images.

Screenshot of Demo Application

How To Use The Library

The following example is taken from the demo application's source code and shows how a bitmap that is locally similar to the input bitmap can be generated.

/// <summary>
/// Runs WFC with overlapping model in a thread creating a <see cref="Bitmap"/> that is based
/// on the specified input.
/// </summary>
/// <param name="inputBitmap">The input image.</param>
/// <param name="options">The overlapping model options.</param>
/// <param name="seed">The seed for the random number generator.</param>
/// <returns>The resulting image or <c>null</c>.</returns>
private static Task<Bitmap> RunWfcAsync(Bitmap inputBitmap, OverlappingWfcOptions options, int seed)
{
    return Task.Run<Bitmap>(() =>
    {
        // Fetch the pixels from the input image (convert Colors to ARGB for speed) and store them
        // in an Array2D.
        var inputColors = new Array2D<int>((uint)inputBitmap.Height, (uint)inputBitmap.Width);
        for (uint x = 0; x < inputColors.Width; x++)
            for (uint y = 0; y < inputColors.Height; y++)
                inputColors[y, x] = inputBitmap.GetPixel((int)x, (int)y).ToArgb();

        // Create WFC overlapping model instance with the created array, the options and the seed
        // for the random number generator.
        var wfc = new OverlappingWfc<int>(inputColors, options, seed);

        // Run the WFC algorithm. The result is an Array2D with the result pixels/colors. Return value
        // is null, if the WFC failed to create a solution without contradictions. In this case one
        // should change the settings or try again with a different seed for the random number generator.
        var result = wfc.Run();

        // Failed...
        if (result == null)
            return null;

        // Success: extract pixels/colors and put them into an image.
        var resultBitmap = new Bitmap((int)result.Width, (int)result.Height);
        for (uint x = 0; x < result.Width; x++)
            for (uint y = 0; y < result.Height; y++)
                resultBitmap.SetPixel((int)x, (int)y, Color.FromArgb(result[y, x]));

        return resultBitmap;
    });
}