banner_image ×
SeaArt AI Empresa
article cover

Creating Complex Poses for Your Characters

avatar
G
avatar_frame
GrayMan
Atualizado em Aug 31, 2025
519

After creating characters, giving them different poses can improve look of your subjects and overall composition. It can be considered as one of the most important side elements of image generation. But there is no term for every poses so we have to connect certain terms to create completed pose for our creations. This allow us to create any pose that comes to mind. But how can we make it effecient and easy, this guide will help you to understand basics of posing and creating your desired poses for characters that you made.

Let's say this is my character and i want to give her different poses for final art.

First thing to do is, of course, thinking about pose. After that we can think about tags or if we have no idea, we can check Danbooru tags for help. I will show a few example to make it clear for beginners. Otherwise it's already easy subject to do guide over it.

For example, i want my character to take mirror selfie because i want to create an Instagram account for her and manage it. For this i need to find my main tag for composition, which is `mirror selfie` for this image. When i search mirror selfie, i can find that tag on Danbooru, that means it's better to use this tag. But using one tag for entire composition can increase randomness of image. If we include wrong generations to this, there is small chance to get what you want in a few attempts if you are not lucky.

But there is related tags section in Danbooru to see what tags used with your tag mostly. You can take pose related ones to improve your character's pose accuracy.

I preferred these tags for my subject for example `looking at phone, holding phone, full-length mirror`. Now my subject has a consistent pose as base to use.

After that, i can use other supportive tags to edit my base pose such as `v`for her hand pose and `head tilt` + `one eye closed` for her head. I can keep editing her pose with changing supportive tags without touching my base pose tags. Which is `mirror selfie, looking at phone, holding phone, full-length mirror` for this subject.

Then there is one problem left. How to not mess the creation and create bad results like the right image. It's simple actually. We have supportive tags but they should not be unnecessary, for example

Left image:

`masterpiece, best quality, amazing quality, newest, very aesthetic, solo, 1girl, red hair, long hair, ponytail, green eyes, black shirt, yoga pants, standing split, holding cup, orange juice, looking at viewer`

Right image:

`masterpiece, best quality, amazing quality, newest, very aesthetic, solo, 1girl, red hair, long hair, ponytail, green eyes, black shirt, yoga pants, standing split, flexible, holding own foot, foot on head, holding cup, orange juice, looking at viewer`

As you can see only difference is supportive tags and it create huge mess in creation. This is because `standing split` already make character hold her leg and keep it up above her head. If we add `holding own foot, foot on head` like unnecessary tags again, it can make AI create this type wrong generations. She already does standing split and now, she does hold another leg above her head and hold it. Because AI just follows orders we have given to it. So be sure that your supportive tags are not just duplicating the pose again and add something new to pose.

Additionally, keep it anatomically possible, for example, if she has 2 hands, don't add 3 tag that uses her hands.

Now only limit is your imagination and capabilities of model you use.

Crazy Samurai

Positive Prompt:

`masterpiece, best quality, amazing quality, newest, very aesthetic, solo, 1girl, black hair, short hair, blue eyes, yukata, front view, standing on one leg, knee up, feet focus, barefoot, holding katana, holding ?, un?ing katana, head tilt, crazy eyes, crazy smile, dark aura, looking at viewer, absurdres, highres`

Negative Prompt:

`lowres, bad quality, worst quality, bad hands, bad ?, sketch, jpeg artifacts, ugly, poorly drawn, censor,blurry, watermark, artistic failure, artistic error, bad proportions, bad perspective, displeasing, very displeasing, oldest, child, childish, nsfw`

Base Pose Tags: standing on one leg, un?ing katana

Other tags are support to our base tags and they improve her look in composition.

160
519
comentário(s)
519
160
0
0/400
Artigos relacionados
cover
Multi-modal posting is now fully open!
avatar
S
avatar_frame
SeaArt Official
62
127
cover
OC Creator Project Excellent Works Collection
avatar
C
avatar_frame
Community Bot-Seira
611
1248
cover
【OC Community Headline】2025 Dec WEEK 1
avatar
C
avatar_frame
Community Bot-Seira
132
528
cover
COMPREHENSIVE GUIDE DIRECTORY
avatar
S
avatar_frame
SeaArt Official
12078
4898
cover
SeaArt Jump Dynamic Cinematography & VFX Creation Guide
avatar
S
avatar_frame
SeaArt Official
5002
1650
cover
curious how to make a long take video?
avatar
W
avatar_frame
wiz.X
4449
2719
cover
SeaArt Dynamic Video Tutorial
avatar
C
chengxu01
186
25065
cover
SeaArt Realism User Guide
avatar
D
DL
667
17886
logo
Português
Aplicativo
Criar imagem Personagem IA Swift AI Treinamento de modelo Canvas Apps de IA Fluxo de trabalho
Sobre
Lab Classificação Chat IA Blog Notícias
Ajuda
Guias Atendimento ao Cliente
Obter aplicativo
icon
Download on the
APP Store
icon
GET IT ON
Google Play
Siga-nos
iconiconiconiconiconiconiconicon
© 2025 SeaArt, Inc.
Copyright Policy
Termos
Política de Privacidade 特定商取引法 資金決済法に基づく表示
Mais