Discover and read the best of Twitter Threads about #stabledifusion
Most recents (7)
A🧵!
In a sea of AI art posts, how do you stand out?
One way, is by adding simple elements of storytelling to your posts in order to encourage the imaginations of those that appreciate your art.
Here are 7 ways to be a better story teller while showcasing your generations:
In a sea of AI art posts, how do you stand out?
One way, is by adding simple elements of storytelling to your posts in order to encourage the imaginations of those that appreciate your art.
Here are 7 ways to be a better story teller while showcasing your generations:
Timeless Wisdom! ⏳
"Show don’t tell". What better avenue to do that in than ai art?
But don’t ruin it by adding a line of context like:
*The “person” did “the action” in order to prevent the “bad thing” from happening.*
Describe the action occurring, not what was done.
"Show don’t tell". What better avenue to do that in than ai art?
But don’t ruin it by adding a line of context like:
*The “person” did “the action” in order to prevent the “bad thing” from happening.*
Describe the action occurring, not what was done.
oooo Shiny! đź’Ž
Bring attention to elements in your image with mystery/intrigue surrounding them. Prompt for “a red sigil”, “a tattoo”, “a facial scar” then mention it without strong context/info.
-sigil glowed brighter
-tattoo brings memories
-scar hurts when raining.
Bring attention to elements in your image with mystery/intrigue surrounding them. Prompt for “a red sigil”, “a tattoo”, “a facial scar” then mention it without strong context/info.
-sigil glowed brighter
-tattoo brings memories
-scar hurts when raining.
I will try to visualise one song with AI, let's see how it goes.
I'll update this thread 🧵
(1/n)
I'll update this thread 🧵
(1/n)
Obstacles 1 Already : Stable Diffusion not starting with custom model.
I'll try for few times if it doesn't work I'll use default model.
(2/n)
I'll try for few times if it doesn't work I'll use default model.
(2/n)
Stable Diffusion is not even starting, forget about creating something.
Curr Goal: Get this thing to start.
(3/n)
Curr Goal: Get this thing to start.
(3/n)
1/10 kicking off our #octaneX news at yesterday’s #AppleEvent - I want to share a preview of a major topic I will cover in my @solana $rndr breakpoint talk in a few weeks:
2/10 On iPad and iPhone, millions of new artists will need to be onboarded to #rndr with a much simpler workflow than desktop 3D tools currently offer. With #RNDR baked into the design of our new app, we believe have a way to address this:
3/10 In March at #GTC22, I first previewed our Neural Rendering work: AI and raytracing seamlessly blend in a single GPU rendering pipeline: #RNDR twitter.com/i/web/status/1…
(THREAD)
Depuis 18 mois le #Web3 est le buzz word qui résulte de l’intersection de 2 tendances : #VR & #blockchain.
Mais la réelle révolution est à l’intersection d’une 3ème tendance : l’#IA générative.
Regardez ces exemples 👇
Depuis 18 mois le #Web3 est le buzz word qui résulte de l’intersection de 2 tendances : #VR & #blockchain.
Mais la réelle révolution est à l’intersection d’une 3ème tendance : l’#IA générative.
Regardez ces exemples 👇
Cette première vidéo est créée à partir d’un simple texte grâce à l’IA générative de Meta.
« Un bébé paresseux avec un bonnet tricoté orange essaie de comprendre comment utiliser un ordinateur. gros plan, très détaillé, éclairage de studio, écran se reflétant dans son œil.mp4 »
« Un bébé paresseux avec un bonnet tricoté orange essaie de comprendre comment utiliser un ordinateur. gros plan, très détaillé, éclairage de studio, écran se reflétant dans son œil.mp4 »
Ces univers 3D immersifs sont générés en temps réel à partir d’images imaginées par une IA #stableDifusion #unreal #psygan twitter.com/i/web/status/1…
Combine #stableDifusion with style transfer models #VToonify and #DualStyleGAN. Toonify videos using #VToonify with the backbone #DualStyleGAN trained on face images generated by #stableDifusion. (1/5) #aiartist #deeplearning
First, generate random art images with #stableDifusion with prompt `A ultradetailed bust portrait painting of a <woman|man>, highly detailed, a full front face in the middle of canvas, upper body, digital painting, sharp focus, _adjective_ mood, illustration by #Artist` (2/5)
Next, train image style transfer model #DualStyleGAN with the these images. The results look pretty impressive! (3/5)
🎉New preprint! We used Diffusion Models (same one from #dalle2 and @StableDiffusion ) to generate 3D MRI images of the brain conditioned on several covariates and make 100k synthetic brains openly available
Paper: arxiv.org/abs/2209.07162
Dataset: tinyurl.com/32p4hu7d
1/n
Paper: arxiv.org/abs/2209.07162
Dataset: tinyurl.com/32p4hu7d
1/n
with @jessdafflon @pfdacosta @PTudosiu Virginia Fernandez, Parashkev Nachev, Sebastien Ourselin
@mjorgecardoso
In a collaboration with @NVIDIA Cambridge-1 supercomputer
2/n
@mjorgecardoso
In a collaboration with @NVIDIA Cambridge-1 supercomputer
2/n
We adapted latent diffusion models (arxiv.org/abs/2112.10752 developed by @robrombach and @andi_blatt) to learn to generate 3D high-resolution medical images. Due to its scalability, we were to train these models on images with millions of voxels!
3/n
3/n
ÂżCĂłmo es posible que #StableDiffusion haya sido desarrollado por un grupo medio desconocido que comparado con los gigantes de la IA (openai, google, meta, etc.) es minĂşsculo?
La sensaciĂłn de "no es posible" recuerda a la irrupciĂłn del open-source (especialmente linux).
Hilo 👇🏻
La sensaciĂłn de "no es posible" recuerda a la irrupciĂłn del open-source (especialmente linux).
Hilo 👇🏻
Una de las máximas hasta ahora es que los modelos gigantescos solo pueden ser entrenados con muchĂsimos recursos que solo tienen los gigantes tecnolĂłgicos.
Esos gigantes controlaran esa tecnologĂa clave y de ahi emanan intuiciones sobre regulaciĂłn, centralizaciĂłn, etc.
Esos gigantes controlaran esa tecnologĂa clave y de ahi emanan intuiciones sobre regulaciĂłn, centralizaciĂłn, etc.
Publicaron estudios para estudiar el camino más óptimos a la rentabilidad y mejores prestaciones de los modelos grandes. Tanto Meta publicó arxiv.org/pdf/2208.08489… y OpenAI arxiv.org/pdf/2001.08361…
