Well, can you think of any algorithm that is able to decide whether a part of an image is tileable?
A simple approach a la "the leftmost columns of pixels is identical (or reasonably close) to the rightmost columns of pixels" does fail on a 'simple' example like a checkerboard.
No, you're quite right. This is precisely what I was thinking about. My question now, I guess, is if anyone is able to produce a script like this? That was my original request, but I forgot to note it in my original post.
...
Well, can you think of any
Well, can you think of any algorithm that is able to decide whether a part of an image is tileable?
A simple approach a la "the leftmost columns of pixels is identical (or reasonably close) to the rightmost columns of pixels" does fail on a 'simple' example like a checkerboard.
Am I missing something obvious?
No, you're quite right. This