テストLORA。
キャラクターの構図や背景が固定されやすくて不自由です。
箒で飛ぶことは、百回に1回成功するかどうかです。
キキ以外はおまけです。
Test LORA.
Character compositions and backgrounds tend to be fixed, which is inconvenient.
Flying on a broom is a one-in-a-hundred success.
Everything except Kiki is a bonus.
*********************************************
アニメLORAですが、アニメモデルとの相性は良くありません、"kantanmix"以外のモデルで出力する場合は重量を下げ、プロンプトで補正したりする事が必要です。
As for the anime LORA, it is not compatible with anime models. If you want to output with a model other than "kantanmix", you will need to lower the weight and correct it with prompts.
*********************************************
好きな衣装や、背景、ポーズを組み合わせて下さい。
Please combine your favorite costumes, backgrounds, and poses.
*********************************************
weight 0.7、ADetailerを使用しface_yolov8nで顔修正を行っています。
高解像度補助:R-ESRGAN 4x+ Anime6B,倍率2倍
Weight 0.7 ADetailer is used and face_yolov8n is used for face correction.
Hires.fix:R-ESRGAN 4x+ Anime6B,2x magnification
*********************************************
"Detail Slider - LoRA"
https://civarchive.com/models/153562?modelVersionId=171989
"Detail Tweaker LoRA (细节调整LoRA)"
https://civarchive.com/models/58390?modelVersionId=62833
"flatBG"
https://civarchive.com/models/81298/flatbg
"noline"
https://civarchive.com/models/81309/noline
"boldline"
https://civarchive.com/models/81313/boldline
"Zoom Slider - LoRA"
https://civarchive.com/models/114460?modelVersionId=123732
*********************************************
"ADetailer"
(インストールは自己責任でお願いします)(Please install at your own risk.)
顔の詳細を修正する追加機能
Additional features to correct facial details
https://github.com/Bing-su/adetailer
*********************************************
Embeddings
"Fast Negative Embedding (+ FastNegativeV2)"
https://civarchive.com/models/71961?modelVersionId=94057
"EasyNegative"
https://civarchive.com/models/7808/easynegative
"negative_hand Negative Embedding"
https://civarchive.com/models/56519/negativehand-negative-embedding
Description
テスト
FAQ
Comments (15)
これでも、何回もやり直して一番ましなのをアップしました。
Very good, I like that the other characters are in there too :)
We worked very hard to increase the number of characters, but it's difficult to have them all appear at the same time.
If you like it, please rate it with ★.
とても苦労してキャラクターを増やしましたが、同時に登場させるのは困難です。
気に入って頂けたのなら★で評価をお願いします。
"Flying on a broom is a one-in-a-hundred success. "Had the same problem with all the Kiki loras.... and all the "riding" prompts for bikes, horses etc. SD just doesn't get the concept.
I tried various methods to make the child learn how to ride a broomstick, but none of them worked.
I think SD learns incorrectly because it doesn't recognize orientation.
If you can prepare a proper four-sided diagram and manage each as a separate trigger, you may be able to remember it.
様々な、箒に乗った姿を、様々な方法で学習させましたが、上手く行きませんでした。
SDは、向きを認識しないので、誤学習するのだと思います。
きちんとした四面図を用意して、それぞれ別のトリガーとして管理出来れば、覚えるのかもしれません。
@FrozenGunner
I think this is one way to output an image of someone riding a broom.
It's a little different from what I want to do, which is learn the concept, because I'm just assigning it in the pose of the control net.
箒に乗っている画像を出力する、手段の一つだと思います。
コントロールネットのポーズで割り当てているだけなので、私がやりたい、概念を学習するのとは少し違います。
@FrozenGunner
I created a LORA that examines the reason why Kiki can't ride a broom.
キキが箒に乗れない原因を考察したLORAを作成してみました。
Is it compatible with NSFW ?
Since it has not been trained, it is not supported by LORA.
With SD, it is possible to output any image depending on the model's capabilities, but we do not recommend it.
学習させていないのでLORAとしては対応していません。
SDでは、モデルの能力次第でどんな画像も出力する事は可能ですが、推奨はしていません。
@jibunsagasinotabi Ok I understand thx !
Fantastic model - thanks!
Any chance of getting SDXL (or pony, either one) versions of these Ghibli characters?
Also, if like to say that I really love all these classic anime/manga styles you've got, like Gunsmith Cats and Mezzo Forte. I especially love that they're flexible enough with sfw art (some pony loras seen to be less well trained on clothing)
Due to PC specs, XL and pony cannot learn freely like SD1.5.
Since calculations and memory are omitted, reproducibility is reduced.
However, the freedom of composition is greatly improved, so I am currently using pony.
@jibunsagasinotabi What are your thoughts on using the Little Nemo characters in the style of the 1984 pilot? I saw model sheets of Nemo, Camille, Flip, Nemo's Dad, Doctor Pill, and Slivers (the last character being seen in the pilot, but is referred to as Omen in the production notes), all these characters were in the original comics, and I find the Ghibli artstyle fitting Nemo a lot more than what we got in the finalized product.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.