Was on some partially conscious shenanigans last night that went about prompting an image, then using that prompted image as a canny controlnet for the same or slightly tweaked prompt, and somehow produced pic related. I can't quite seem to figure out what actually creates this kind of analog depth anime effect though. Messing with the quality tags seems to have some effect but I can't seem to get it consistently. But it sporadically spits out images like this
https://files.catbox.moe/0vasvi.png where it really starts to feel more like an anime illustration/screenshot than an AI gen. More messing around is in order.
