|
|
|
|
|
|
by wang_li
320 days ago
|
|
|
There is a naive approach to making this kind of thing that reduces the component images to such a small size (2-3 pixels in large image) that turns this into more of a dithering exercise than looking for artifacts in each component image to match up lines. It's still a nice effect, but it's quite different when the component images are > 10% the size of the final image, instead of < 1% the size. |
|
|