CivArchive
    Monster FannyPacking - ControlNet Pose Pack - Monster Fannypacking
    NSFW
    Preview 3206230
    Preview 3206232
    Preview 3206236
    Preview 3206237
    Preview 3206229
    Preview 3206231
    Preview 3206235
    Preview 3206234
    Preview 3206238
    Preview 3206239
    Preview 3206240
    Preview 3206241
    Preview 3207150
    Preview 3207259

    So uhhhh.....based on request. Because this is a thing, I guess.

    Fannypacking is, and I quote, "the concept of a woman tied to a monster's chest as a human armor" - so TIL. And here's a pose pack for it.

    Good results seem to come from pairing OpenPose + Canny images, check out my previews for prompts.

    Description

    FAQ

    Comments (13)

    NextMealOct 27, 2023
    CivitAI

    Hi, is there a recommended weight for each controlnet? Could you kindly attach some results? Ive done something similar to this, maybe a little more complex, but couldn’t really get usable results.

    Akt1887
    Author
    Oct 27, 2023· 1 reaction

    I'm guessing you're gonna need some regional prompting + inpainting to make this really work, and maybe a LoRA if one exists. The poses are generally no problem with ControlNet, its the whole "strapped to orc's chest" thing that fucks it up. I'm getting some results though, check back in a bit and I'll upload some samples.

    NextMealOct 27, 2023

    @akti88 Ok, I see. Thanks for the suggestion.

    Akt1887
    Author
    Oct 27, 2023

    @NextMeal Oh dude I just figured out who you are, I have a bunch of your sliders, great work man love your stuff

    NextMealOct 27, 2023

    @akti88 Thanks mate 😊

    bitzupaOct 27, 2023· 1 reaction

    can be helpful sometimes but sometimes it makes mess even when using dedicated lora. There are several issues i noticed, most of your poses have monster legs out of picture resulting in SD generating monster with no legs or standing on knees or replacing monster legs with girls legs... You also should add some background wall or something as pose alone will simply not work is too complex for SD but it will work with pose+depth but using depth of just characters results in one color simple background. Will post generation of one working pose so you see what i mean and then will post same thing but using lora and and simply picture of such concept as depth input. Was testing openpose and depth that can be generated from posemyart etc and i must say its finnicky as hell so i figured out way to get almost all poses (excpet for side poses from some reason) by using DAZ3d curated renders as input for canny + depth and it works like charm, meaning posemyart stuff is simply lacking info for proper generation. That's why i would suggest adding backgrounds, floors etc so it has easier time actually capturing depth and whats what.

    Akt1887
    Author
    Oct 28, 2023

    @bitzupa Thanks for the tips and experiments man, I'll try to play more with PMA's depth, it has a slider and a positional lighting thing so I can mess more with it, good advice also about adding a wall

    SmallmiOct 27, 2023
    CivitAI

    It's still hard to present "Orc" as a green-skinned orc.

    it's a lot time will turn into a normal male.
    sad pepe

    bitzupaOct 27, 2023

    use orc lora, use monster in prompts, use one of funny packing loras tthey are biased toward monsters.

    pico0obyte442Oct 28, 2023· 1 reaction

    the regional prompter might help

    SmallmiOct 28, 2023

    @bitzupa I think the character models are too close to humans rather than monsters.

    SmallmiOct 28, 2023

    @pico0obyte442 cool plugins , thank you dude!

    Poses
    SD 1.5

    Details

    Downloads
    566
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/27/2023
    Updated
    5/13/2026
    Deleted
    -

    Files

    monsterFannypacking_monsterFannypacking.zip