Discover and read the best of Twitter Threads about #Dreambooth

Most recents (24)

OK! Finally dropping my new #DreamBooth fork--you can now use BLIP to caption your dataset, then extract image captions to txt files. This lets you train on the text/image pairs instead of providing general instance prompts.
#aiart #stablediffusion…
You can ALSO specify instance and class prompts to train on *alongside* the image captions. they get appended to the image caption in the text file.
You can train multiple concepts at once, too! I've been getting some next-level results out of it so far.
LMK how it works for y'all!
Read 4 tweets
I tried 25+ Generative AI API platforms so you don't have to

And I think I found the best one and it's mind blowing!

Here it is (Free 5000 API credits):

On most platforms, Developers like me have to deal with:

High cost

Headache of managing GPUs

Scaling issues

Monster API resolves such issues by providing easy-to-use, scalable, and affordable APIs for developers to access generative AI models
With Monster API I was able to effortlessly access powerful AI models which auto-scale based on the requirements

Developer-friendly APIs made it super easy to access these models in multiple tech stacks
Read 10 tweets
Do you want to quickly turn pre-trained image models like #StableDiffusion and #DreamBooth into a text-to-video / GIF generation model? Excited to share our #TuneAVideo takes only 5-10mins on 1 A100 for training and is now open sourced:… (~500 stars so far)
Joint work with @jayzhangjiewu @YuchaoGu @ge_yixiao @_Xintao_ and we welcome further discussions and collaborations!
Also shoutout to @huggingface @AK for creating a great model zoo library for #TuneAVideo models conceptualized and fine-tuned by Community at…
Read 3 tweets
🎉Introducing RoentGen, a generative vision-language foundation model based on #StableDiffusion, fine-tuned on a large chest x-ray and radiology report dataset, and controllable through text prompts!

@PierreChambon6 @Dr_ASChaudhari @curtlanglotz

🧵#Radiology #AI #StanfordAIMI Text-conditioned synthesis of CXR. Each image was hand-picke
#RoentGen is able to generate a wide variety of radiological chest x-ray (CXR) findings with fidelity and high level of detail. Of note, this is without being explicitly trained on class labels. Synthetic images created by prompting a fine-tuned model for
Built on previous work, #RoentGen is a fine-tuned latent diffusion model based on #StableDiffusion. Free-form medical text prompts are used to condition a denoising process, resulting in high-fidelity yet diverse CXR, improving on a typical limitation of GAN-based methods.
Read 13 tweets
121 little characters with pointy ears. A small village.

All 100% designed with #AI.

Men, women, kids, soldiers, wizards, goblins, etc. All created in less than one hour, while keeping a solid style consistency.

Thread 🧵

#StableDiffusion #Dreambooth #Gaming @Scenario_gg
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.
Read 15 tweets
Stop everything you're doing and just use "plasticine" in your prompts lol.

New #goblin model coming.

#StableDiffuson #Dreambooth #AI ImageImage
Read 10 tweets

I've made some gems too. 100% #AI-generated 🤯

Sorry, I don't have much time for a detailed thread this time... too much work atm. I just wanted to put this out there and show it's really working for SO many game assets!

#StableDiffusion #Dreambooth #Scenario #AI
This is actually a fun one, I'll try sharing the model when I get a chance.
Read 11 tweets
¿Te gustaría animar una imagen generada por #stablediffusion – incluso con #Dreambooth – con un video grabado de ti mismo? Aquí tienes otro ejemplo 👉

Pues sigue los pasos que te dejo a continuación que no son pocos :) Let's go! 🧵👇
1/ Carpeta para el proyecto.
2/ Grábate un clip de video con webcam + OBS o tu app favorito. Lo + importante:
- Mínimo HD.
- Cámara que se mueva despacio.
- Que no aparezcan elementos nuevos en el plano: objetos o partes de cuerpo.
- Bien iluminado
- 10 seg. de largo aprox…
3/ Escala a 512px de alto y luego recorta tu video a 512x512px con tu app favorita. También puedes hacerlo online con
3.1/ Guarda el video resultante en tu carpeta de proyecto. Escalar el video a 512px de...Recortar el video a 512x512...
Read 9 tweets
Il y a 5 mois je vous présentais une des premières IA disponible pour le grand public pour créer toutes les images que vous voulez.

En 5 mois, beaucoup de choses ont évolué, aujourd’hui je vais vous présenter #Dreambooth
On retrouve maintenant un peu de concurrence, d’un côté des acteurs « privés » avec des services en ligne (#dalle2, #Midjourney) de l’autre côté des acteurs « ouverts » comme #StableDiffusion…
L’un des avantages d’une technologie ouverte, c’est qu’elle peut être reprise par des dizaines d’autres chercheurs, améliorées, et c’est ainsi que StableDiffusion grandit aujourd’hui.…
Read 19 tweets
I've done a first pass on the dwarves, and they're fun.

This proves again that #AI / #StableDiffusion is a fantastic tool for designing and creating #game assets... when adequately fine-tuned.

Here's some of the exploration, so far - 🧵
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"👇
Read 16 tweets
Sharing additional thoughts on #StableDiffusion to create #gaming content & #game assets 🧵

My last exploration was on golems; this one is with "Space Marines-like" heavy infantry.

This was a fun creation with lots of learnings 👨‍💻. Feel free to like/RT if you find it useful 🚀
1/ First (as mentioned before - ) > curate a training dataset (such as pictures of figurines) to feed a #Dreambooth finetune.

Once the model is ready, compare different generic prompts (e.g. "low-poly," "3D rendering“, "ultra-detailed," "pixel art," etc.)
2/ Once a prompt looks good, keep the modifiers (in this case: "3D rendering, highly detailed, trending on Artstation").

And start iterating around variations, such as colors (or pose). Don't over-engineer it, to keep the consistency. You should get this:
Read 15 tweets
Songs of Wander

#ai #street #photography ImageImageImageImage
Songs of Wander 2

#stablediffusion #dreambooth ImageImageImageImage
Songs of Wander 3

Feel like losing my mind a bit, getting lost in creating these images. But it's a fine feeling. ImageImageImageImage
Read 4 tweets
#Dreambooth will change how 2D game assets are being designed or generated.

Today's exploration: treasure chests 💎💰.

I trained Stable Diffusion with some low-poly treasure chests (23 images). 1h later, here's what's possible (🧵): ImageImageImageImage
At first, I randomly generated a set of 64 items, without any specific prompt engineering ("a treasure chest"). Image
Here's a close-up view: Image
Read 11 tweets
Tonight's dreamboothing is about grenades 💣(!).

I made a #dreambooth finetune from a few Sci-Fi grenade illustrations.

Explosive thread 🧵👇

#AI #StableDiffusion ImageImageImageImage
First, some random grenades, no specific adjectives. Image
But they look great in blue too! ImageImageImageImage
Read 14 tweets
I designed 160 weapons for a heroic fantasy game, inspired by World of Warcraft... in just a few minutes.

A short thread on Generative AI & gaming 🧵

#GenerativeAI #AI #StableDiffusion #Dreambooth #Midjourney #Dalle #Gaming #UGC
Generative AI is likely to go big in vertical communities such as gaming, per @Tocelot - “The fringe is where revolutions start”.

AI can help design and quickly create new gaming environments, textures, scenes, characters, weapons, skins... or props.

@Tocelot Dalle, Midjourney or StableDiffusion generate exquisite images. However, it’s challenging to reconstruct specific subjects while maintaining high fidelity to their key visual features. Even w dozens of detailed iterations, the outcome is rarely consistent. Prompting is daunting…
Read 8 tweets
He seguido el tutorial de @DotCSV para entrenar a la IA para que cree imágenes con mi cara y...

#IA #StableDiffusionART #Dreambooth ImageImageImageImage
Read 18 tweets
Dreambooth training for Stable Diffusion is extremely powerful. You can train a new token to the "person" class to create very convincing looking images of them. I've posted some examples in the past days.

But it's not the coolest thing you can do...
You can train the "style" class to create new styles. For instance the "arcane" style is well known - it skews all results to this particular Riot games style.

But - that's not the coolest thing you can do...
It's ridiculous, but you can just train the "style" class to be photographic. And you get back results that - well, look photographic.


Read 10 tweets
Black portraits

Dreambooth training SD 1.5 model for photographic style.

#dreambooth #stablediffusion #portrait
Read 5 tweets
Andy Griffith (Matlock era) in...

Top Gun / Avengers / Star Wars / Game of Thrones

#andygriffith #dreambooth #stablediffusion
Andy Griffith (Matlock era) in...

Mad Max Fury Road / Pirates of the Caribbean / Gone with the Wind / Star Trek

#andygriffith #dreambooth #stablediffusion
Andy Griffith in Spider-Man

It's pretty incredible what's happening here. By prompting Andy, and using the Dreambooth model, I get back an old Spider-Man. It's him in there!

#andygriffith #spiderman #dreambooth #stablediffusion
Read 4 tweets
#Dreambooth is a method to teach new concepts to #stablediffusion , we have a super simple script to train dreambooth in 🧨diffusers. But our users reported that the results weren't as good as other Compvis forks. So we dug deep and found out some cool tricks.
A 🧵
Training the text encoder along with the unet gives the best results in terms of image-text alignment and prompt composition.

Left image - frozen text encoder
Right Image - finetuned text encoder

The results are drastically improved :🤯 ImageImage
We updated our script to allow fine-tuning text encoder…
Read 11 tweets
AI unlocks unlimited product photography. 🤯

Recent advances have made it possible for merchants to turn mobile photos into high quality product images + ad creative.

AI can now learn and reproduce a product on demand. More…👇[1/10]

#ai #aiart #stablediffusion #dreambooth
Products with simple forms and organic lines reproduce exceptionally well. We took this trainer on adventures across unfamiliar terrain. 👟🏔 [2/10]
This mid-century Eames Lounge got a tour of some Bond-villain HQs nestled into the world’s most beautiful landscapes. 🏜🧊 [3/10]
Read 10 tweets
GM! Here's the daily summary of trending projects on crypto GitHub. Curated by the greatest web3 developers ⭐️.

Check the 15 projects in the thread below 🧵👇🏻
0xYYY and 6 others starred repo EYBlockchain/starlight (79 ⭐️)

:zap: solidity --> zApp transpiler :zap:…
0xYYY and 5 others starred repo libevm/ (14 ⭐️)

Simple no nonsense tooling for Optimism…
Read 15 tweets
Muchos (no han sido tantos) me habéis preguntado cómo entrenar a la IA con tu cara.

Voy a tratar de explicarlo en este hilo.

Todo el crédito a TheLastBen por la creación del Colab y por su ayuda.
1⃣ Crea una cuenta en (es gratis) y confirma el email.

En tu perfil, ve a Settings > Access token > New Token: ponle el nombre que quieras y en Role elige Write.

Copia el token.
2⃣ Abre el Colab y crea una Copia: Archivo > Guardar Copia en Drive.…

Desde tu copia ejecuta las 3 primeras celdas en orden. Dale a ▶️ y cuando veas el✔️, ejecuta la siguiente.

Cuando ejecutes la 1ª celda te pedirá conectar y acceso a tu drive. Si a todo.
Read 11 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!