The model was trained on just 11 pictures (!), with only 1500 training steps, which tuned out to be quick (20 min).
As before, the first step is to "explore“ the model with a few generic prompts. The goal is to find the modifiers that will keep a consistent style going forward.
Once the "stable modifiers" are found, it's time to select some of the best output and remove the background when needed.
"A dwarf, detailed, trending on Artstation, Clash of Clans"👇
You can add accessories or pose, for example: "A dwarf lord sitting on a throne". You get images with the throne... and the crown.
This is a group of dwarves that bear an axe 🪓 (... supposedly - since none of the pictures in the training dataset had an axe, I'm not getting the best results here)
Here's another group, this time bearing "a sword and a shield". A little bit better.
Here's a group of baby dwarves, on a baby chair
You can play with expressions too. Include "laughing" in the prompt, and you'll get characters like the ones below (male or female, btw)
Of course, I had to try my zombie trick: "a zombie dwarf, trending on Artstation, clash of clans".
Zombify everything.
I made some elves... in the style of the dwarves
Let's not forget img2img. this was a test to reproduce Russel Crowe's famous moment in Gladiator... "Are you not entertained?"
You can change the colors of the clothes simply by adjusting the prompt: green, blue, red... or bare chest
And then there's always the "fun and creative" stuff like having your characters play cricket 🏏, basketball 🏀, curling 🥌 or bobsled 🛷
This is not just about single characters, they can be generated in groups ("Dwarves in a gold mine, detailed, trending on Artstation")
And I could keep going on and on, but some other datasets are waiting :)
While images might not be final, this can accelerate design or prototyping processes. I also see RPG communities using this to build entire worlds "in the style of the game", without erratic prompting.
If you like this thread, please feel free to like/RT or share any thoughts below.
Follow for more explorations on how to use #StableDiffusion and #Dreambooth to accelerate your game creation process 🚀
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.