List view

• Subdivision level for tiles

  Instead of comparing the average color of a whole tile with the average of the area that needs a tile, the tiles shall be subdivided into more areas. finding the image that best fits to most of the sublevel average colors will bring more detail to the resulting image. the drawback is that a lot more images will be needed to find a good fit.

  No due date
  0% complete0 open0 closed

• Patched image clusters

  Instead of only finding the best fit for one tile, there should be an option to find a tuple of images that fit for the average color of neighboring tiles in the case that no images could be found within the tolerance bound. This will lead to better results where otherwise holes would appear in the image.

  No due date
  0% complete0 open0 closed