I bought a pack of characters with pointy ears. They made me think of the goblins in Harry Potter (but more human-like)
I trained a finetune using @Scenario_gg (internal alpha) and started prompting (calling them "persons").
At first, you get this, but then the fun begins.👇
A basic prompt, no modifier ("a person")
All are perfectly consistent (the clothes, the ears, the accessories). There are some variations, though.
For a village, you need males and females, so I did "a female person".
Despite having zero female characters in the initial training data, it generated this:
Any village has kids, so kids I did. Boys and girls.
With pointy ears like their parents.
The girls look even better.
When a specific character looks interesting, it's worth generating some variants (different poses, accessories, expressions).
I just use img2img on the initial character to get 16, 40 or more.
They can dance, too ("a dancing person").
Every village needs protection, so I did some soldiers.
The soldiers could be samurais, too (with pointy ears).
This gets further than the original style, so there are some discrepancies.
Same for "the Inca warriors"
Or the African tribesmen (raw output, without background removal)
I usually finish explorations with fun prompts, for example, using "#plasticine" (newly discovered modifiers).
"A plasticine figurine of a person"
They would be amazing, 3D printed.
That's it!
It's the type of exploration we want people to do, using @Scenario_gg, soon.
If you liked this thread, please feel free to RT the first tweets, like/follow, or just leave your email, so you are notified when we launch🚀 >> scenario.gg
Here are the key steps to creating stunning turnaround, using #Scenario ()
1/ Train or pick a character model (A).
2/ Optionaly>, pick a style model (B). Use it to create training images for (A), or you can merge both (A + B = C) for example.
3/ Utilize the custom model (A or C) to generate consistent characters. Then select a reference image to produce initial character turnarounds in your desired poses.
4/ Refine these initial outputs using Sketching and image2image.
5/ Select the best result and refine details in the Canvas for maximum consistency.
6/ Finally, upscale your final image (up to 8K resolution.)
@ClaireSilver12 - (I hope you don't mind me RT this for a broader reach and to share it with more users.)
Here's an advanced use case for the IP Adapter. You can adjust or remove the steps depending on the desired output/goal. Bear with me; it's actually quite straightforward.
1 - Train a LoRA on a specific subject (e.g., character).
2 - Blend the LoRA to perfectly capture the style (e.g., comic, cartoon, oil painting, 3D...).
3 - Run inference on that "blended" model.
4 - Select an image that stands out and use it as a reference with the IP Adapter.
5 - Modify the prompt to create variations of the subject.
Let's get started 👇👇
1/ The first step is to train one (or more LoRA) models on a specific subject (e.g. character or object), or also a style.
The process is straightforward. I'll use the example of the "girl with pink hair" (😊🫠) that I shared before (12 training images)
Simply select "New Model - Train" on . I use 9 images of the model, showcasing various angles and zoom levels, accompanied by concise captions (details below).
This could be the best model I've ever created for generating isometric buildings, on Scenario.
Output consistently match the style I wanted, and the model responds perfectly to (short) prompts, without any reference images needed.
It's a LoRA composition. More below.
Process: it's pretty simple.
I created a LoRA composition from 4⃣ distinct LoRA.
(i) - My own "Fantasy Buildings" LoRA
(ii) - Three LoRAs available on #Scenario: "Isometric Storybook", "Stylized Fantasy Iconic Imagery" and "Belgian School Comics".
The influence of each LoRA is below.
My prompt structure was dead simple... less than 10 words!
(type of building/scene), solid color background, highly detailed, centered.