Sniper TagTeam Human Bipod (Pose)

LORA
Original


Updated:

Sniper TagTeam Human Bipod

Trained on Illustrious-XL.

This lora produces a pose consisting of 2 persons, where one is aiming a rifle/sniper and the other one is acting as a human bipod supporting the shooter.

You can use the provided dataset for anything you like. If you are using it in any publication I would appreciate a reference to this model, but it's not required.

Example Prompt:

5n1per_t4g_te4m, 2 girls, 1girl with sniper rifle, 1girl, long hair, green hair, blue eyes, twintails, aiming, seiza, other girl lying on floor, red eyes, black hair, hips up, lying on lap, rifle on ass,

Prompt Engineering:

The prompt order and descriptions of the persons are essential for this model to give good results.

Note that the character descriptions will not be always applied correctly by the model (Use inpaint).

  • The prompt should be build in the following order

    • Trigger: 5n1per_t4g_te4m, 2 girls

    • Shooter:

      • 1girl with rifle (you can specify the type of rifle her, e.g sniper rifle, assault rifle,..)

      • Optional: give some information about the rifle like:

        • 1girl with assault rifle, ar15

        • 1girl with assault rifle, ak47

        • 1girl with black sniper rifle

      • After that describe the shooters properties:

        • start the character description by saying 1girl/1boy

        • 1girl, long hair, black hair, green eyes, black hoodie, ...

      • Select a shooter sitting position:

        • aiming, sitting

        • aiming, seiza

        • aiming, wariza (does not work well)

      • Once you are done describing the shooter add the bipod prompt:

        • other girl lying on floor, hips up, lying on lap, rifle on ass

        • lying on lap can be removed or extended/replaced by: head between legs, head on back, sitting on head (not all of these might always work well)

      • And again you can follow with the bipods properties:

        • long hair, black hair, green eyes, orange hoodie, ...

Here is a final prompt template (<...> needs to be replaced):

5n1per_t4g_te4m, 2 girls, 1girl with <rifle_type>,<additional_rifle_descriptions>, 1girl,<shooter_description>, aiming, <sitting position>, other girl lying on floor, <human_bipod_description>, <head_position>, rifle on ass, <environment_description> 

For more help look at the showcase images.

Usage Tips:

  • Strength: 0.8 - 1

  • Describe the shooter and bipod as explained

  • Use 0.3 denoising strength when applying HighResFix

  • I don't recommend a denoising strength >0.4

  • Use Landscape resolutions for best results (also works with portrait)

  • Use background and environment prompts, also describe the character for more variety

  • To get specific different characters you might wanna use inpaint

Resolutions I use for good results:

  • 1024x1024

  • 1152x896 (Landscape)

  • 1536x1024 (Landscape)

Usually I also apply a 1.5xHighResFix (0.3-0.5 denoising strength)


I also love to see the images you create using my loras so I would appreciate any post.

Version Detail

Illustrious

Project Permissions

    Use Permissions

  • Use in TENSOR Online

  • As a online training base model on TENSOR

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated contents

  • Use on generation services

  • Sell this model or merges

Related Posts