Details
Related
v1.0
Civitai 銅像化 / Statue transformation

Civitai 銅像化 / Statue transformation

36
13
85
#concept
#statue
#transformation
#Illustrious

銅像化コンセプトLora


プロンプト調整制御が非常にシビア。
ついでに外部Loraキャラだとかなり厳しい。

良くてモデル内蔵キャラが限界かな。そもそも出来微妙だが。


■推奨強度
0.6-0.8 内蔵モデル利用時。
1.2 外部Loraキャラ使う場合。但し、そもそも外部Loraキャラ利用を推奨出来ない
Adetailer推奨したいが、制御が面倒

枚数多く生成する場合、Adetailer使わずに作ってから構図いいものを後からAdetailerで再生成した方がマシ(複数銅像が出やすいため)

人間っぽく見えるのでblushが天敵。blush抑えてもたまに出てくるが。

■注意点

・【重要】たまに目がホラーっぽくなる。

・モデルの銅像タグに伴うものか、複数銅像が発生しやすい(Adetailer処理の長時間化)
・Adetailerで銅像化を部分解除しがちになり気味(Adetailerにも軽度Lora利用する等の抑止は少しは出来るが、いずれにせよ制御がシビア)
・ある意味、当たり前だが服装やヘアスタイルの色指定するとそこだけ色が出るようになる。ちなみに銅像の色ベースは緑。あえて服を浮かせる手もあるが。
 色指定してなくてもたまに服装の色浮くときがある。
・出来自体、微妙。Control net使った方がいい出来栄え。

・そもそもLora使わなくても出ない事は無い。像そのものらしくなったりマネキンっぽくなったりキャラ表現との兼ね合いが微妙かつ、銅像の土台も一緒に表現したかったので学習を選んだが

English


Statue Transformation Concept LoRA

Prompt adjustment control is extremely strict.

Using it with external character LoRAs is very unstable; at best, built-in model characters work, though even then results are only mediocre.


■ Recommended Strength

0.6–0.8 when using built-in model characters.

1.2 if using external character LoRAs (not recommended).

ADetailer is suggested, but handling control is cumbersome.

For bulk generation, it’s better to generate without ADetailer first and then re-render selected good compositions with ADetailer (since multiple statues often appear).

Characters still look “human-like,” so blush is a major problem. Even when suppressed, it sometimes shows up.


■ Notes

Important: Sometimes the eyes look horror-like.

Because of the model’s “statue” tag, multiple statues tend to appear (causing longer ADetailer processing).

ADetailer often partially cancels the statue effect; using light LoRA strength with ADetailer can help, but overall control remains severe.

Naturally, if you specify colors for clothing or hairstyle, only those parts will appear in color. The statue base color is green. Occasionally, clothing color floats even without ???????? color prompts.

Overall quality is mediocre; using ControlNet yields better results.

In fact, statues can appear without this LoRA at all (sometimes mannequin-like). I trained this LoRA because I wanted both character and statue base depicted together.

View Translation

Rating & Review

-- /5
0 Ratings

Not enough ratings or reviews received yet

B
Chat with the model
Notice
2025-08-24
Publish Model
2025-08-24
Update Model Info
Model Details
Type
LORA
Publish Time
2025-08-24
Base Model
Illustrious
Trigger Words
gdzo,
standing, full body, (solo:1.5),
green statue girl, statue eyes, statue clothes,
(colorless clothes:2),
statue base, (brown base), statue base box,
no human,
Copy
Version Introduction

negative prompt: (blush:2),

License Scope
Model Source: civitai

1. The rights to reposted models belong to original creators.

2. Original creators should contact SeaArt.AI staff through official channels to claim their models. We are committed to protecting every creator's rights. Click to Claim

Creative License Scope
Online Image Generation
Merge
Allow Downloads
Commercial License Scope
Sale or Commercial Use of Generated Images
Resale of Models or Their Sale After Merging
QR Code
Download SeaArt App
Continue your AI creation journey on mobile devices