Convert cubemap image to an equirectangular spherical projection image using Python 2.7.
Currently only works with cubemaps laid out in the following format:
________
| |
| top |
| |
_______|________|_________________
| | | | |
| left | front | right | back |
| | | | |
|_______|________|________|________|
| |
| |
| down |
|________|
- Open an image using PIL or equivalent.
- Calculate dimensions of output image. (Width must be 2xheight. Height can be any percentage of input image, but larger output images lead to less precise results.)
- Calculate length of each square face (ie. 1/3 height of the input, due to height being made from 3 faces - top,front,down)
- Create an output image using the new dimensions.
- For each pixel location in the output image run
cube2equi.find_corresponding_pixel()
to return the corresponding pixel in the input image. - Take the colour value for the input image pixel and apply it to the pixel in the output.
- Automate all setup steps
- Allow for different cubemap layouts