Talk:Wave function collapse: Difference between revisions
m
→Vote for deletion
(hmm...) |
|||
Line 13:
:: Also, ... after thinking about this a bit, it's not clear to me whether "edge of the image" (no adjacent pixels on one side or (for corners) two sides) should be treated as a "tile" (probably not: the other possibility, suggested by "tiling" is that each edge of a tile is connected to the opposite side of the image, meaning that we can select regions of pixels which would appear when repeating the image as a tile). But, ... I think that that kind of detail is an example of information which should go into the task description (if there was a task description). And... this gets back to the lack of any specific task examples presented here. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 07:40, 9 July 2022 (UTC)
:: There's probably also a constraint that the replication tiles (e.g. 3x3 sized tiles) are "aligned" on a grid which matches that tile size. And, given the "tiling" concept, ... conceptually if we were generating an 11x11 image with 3x3 tiles we'd do something like construt a 12x12 image and then crop it to 11x11. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 08:14, 9 July 2022 (UTC)
:: After playing with some implementations, I'm wondering if maybe there's a bit of smoke and mirrors in the implementation at https://github.com/mxgmn/WaveFunctionCollapse#. (I'm getting contradictions -- empty results -- so far for every attempt I've made, running my version of wave function collapse. This suggests that I am imposing a constraint which was not present in the github implementation. I have yet to determine what it is about my interpretation of the algorithm description which differs from that github implementation.)
:: For example, with a starting bitmap of:
::1 1 1 1 1 1 1 1 1 1 1
| |||