Загрузка видео...

Не удалось загрузить видео

На главную

IT'S FINALLY HERE! 🔥 Mystic Structure Reference! 🔥 Generate any image controlling structural integrity ✨ Infinite use cases! Films, 3D, video games, art, interiors, architecture... From cartoon to real, the opposite, or ANYTHING in between! Details & 12 tutorials 🧵👇

311,073 просмотров • 1 год назад •via X (Twitter)

Комментарии: 24

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

Available NOW at Magnific 🪄 for all users! 👇

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

Super easy to find: 👉 Magnific > Mystic > Structure reference Don't forget a good prompt to guide your generation! ℹ️ Currently Style Ref and Structure Ref can't be used at the same time. We are working on that!

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

Let's see some examples! This is a "Anything to anything" meaning you can... 1. Go from cartoon to real "Film still, real actors dressed as Scooby-Doo characters, cinematic light" - Structure strength: 50% - 2K - Creative Detailing: 52%

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

2. Or go from real to cartoon... Is completely up to you! - "As a loose pencil sketch, an open hand with a deck of cards" - Structure strength: 50%

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

3. Or to whatever because why not? xD "Plastic characters, toys, sunny day, cute, beautiful, tropical paradise" (2k resolution - Creative Detailing 52%)

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

👉 So, what exactly is "Mystic's Structure Reference"? Pretty much what the name says: it lets you keep the structure of any image and guide the generation with your prompt. The fun part? Playing around with the Structure Strength slider + prompt can lead to some pretty wild results! 👉 And what IS NOT "Structure Reference"? 1️⃣ It’s not Style Transfer (though soon it’ll work alongside our SREF). 2️⃣ It’s not a direct transformer of your original image. Well… kind of. But it doesn’t see the colors or textures of your original image. That said, we’re also adding a slider soon that will let you do that, along with SREF too, which will replace our current Style Transfer spell and give you LOT OF CONTROL.

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

Kinda amazing, isn't it? 4. "Marble statue, impressive, As created by Michelangelo in the Sistine Chape, classic art" Structure strength: 50% Original image by @sausirhi

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

5. You can bring your sketches to life. "film still, a shot from a fantasy film: an old dwarf sitting at an inn, holding a silver mug. Cinematic lighting, mysterious atmosphere. In the background, a busy dimly lit inn with many customers" - Structure strength: 100% - Plus a 2x Upscales with Magnific Original image by @TheoChron

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

6. User lower Structure Strength for more freedom! "By Quentin Tarantino, Fill still, real man, brown hair, with money bills in his hand, epic scene, impressive, funny, realistic, futurama style but real film, 3d cartoon hyper realistic, hi-res" 25% Str. This one is KABOOM 🤯

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

7. Or a higher Structure Strength to get an almost "colorizer" effect (but keep in mind that it won’t work well with images that have tons of small details due to limitations).

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

8. It also works for pixel art :) "pixel art, 16-bit fighting game by Capcom from the 1990s. background stage should be in the same style with the cherry blossom and building in the background." - 50% Structure - 2k + 52% Creative Detailing

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

9. For texturing basic 3d models / sketches. "Photography of a beautiful house, grass, realistic, detalied" + similar prompts - 50% - 70% Structure - 2k or 4k with Creative Detailing 50% - 85%

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

10. If you feel artistic, you will enjoy a lot too with this tool :) - "Portrait photography of a man, a vintage photo, 1960, artistic" - Structure Strength: 50% Original image by @gizakdag :)

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

"Portrait of a man with a flower shirt, blue eyes, by Pixar, 3d cartoon, sunny day, colorful, realistic, 4k, in the beach, blue sky and palms"

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

11. You can even bring dinosaurs back to life! 😜 "Oil painting by Monet of Tyrannosaurus rex, film still, a dinosaur in a jurassic fern forest, sunny day, near a lake, a volcano in eruption in the back" - Structure only 24% - 2k + Creative Detailing 33%

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

12. Control! That's all we want to give you at Magnific. "Two primitive tribe youths holding spears want to hunt a scary monster in a cave. The sky and surrounding nature are fantasy themed, colorful, mushroom forest, dark cave, pink monster" Original sketch by @iamHadaJZ

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

You have a TON more of examples here:

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

🔴 Limitations: NO, STRUCTURE STRENGTH IS NOT PERFECT. I HATE IT TOO, BUT THAT'S JUST HOW IT IS: If your image has A TON of details Even if you crank Structure Strength to the max... it just won't work. Sorry. We’re not there yet. Maybe in the near future.

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

But come on... It's AMAZING already, isn't it??? "Quentin Tarantino, Fill still, real man Quentin, brown hair, with money bills in his hand, epic scene, impressive, funny, real film, hi-res" 25%

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

Can't wait to see what you will create with this tool!

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

OK, enough! TIME FOR YOU TO ENJOY! GO! GO! GO!

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

Show me your best results!!!

Фото профиля Javi Lopez ⛩️
Javi Lopez ⛩️1 год назад

If the image isn’t coming out how you want, try tweaking the prompt and playing around with Structure Strength in jumps like 25% - 50% - 70%. Sometimes, sticking to 50% isn’t the way to go… but extremes usually aren’t great either. Start at 50% and fine-tune from there!

Фото профиля AndaSeat
AndaSeat1 год назад

✨ Marie Kondo would approve! Phantom 3's clean aesthetic: Sleek, minimalist design Hidden assembly points Magnetic bracket covers No messy cables or complicated parts Does it spark joy? Absolutely! 🎯 Declutter your space: 💰 Minimalist special: Simplify your life with $20 off! #AndaSeat #Minimalism #CleanSetup #Phantom3

Похожие видео

Want to create an avatar from a single image? FlexAvatar is a transformer model that creates full 360°, high-quality, and expressive 3D head avatar from just a single portrait image in minutes. Real-time Demo: FlexAvatar's lightweight architecture allows both animation and rendering in real-time, enabling interactive user experiences. To create a new 3D head avatar, only one image is required, e.g., from a webcam. The final avatar is ready after 2 minutes. Architecture: Under the hood, FlexAvatar adopts a transformer-based encoder-decoder design. The encoder maps the input image onto a latent avatar space, while the decoder produces 3D Gaussian attribute maps by incorporating the animation signal via cross-attention. The model learns all facial animations directly from the data without relying on pre-built 3D face models. This equips the avatars with realistic facial expressions. The internal avatar latent space can be conveniently used to integrate additional observations of a person via fitting. This enables use-cases where more than one image of a person is available, e.g., from a phone scan of the person. We train jointly on 2D monocular videos and multi-view data. However, in monocular videos, the animation signal leaks the target viewpoint, causing the model to produce incomplete 3D heads. We call this phenomenon entanglement of driving signal and target viewpoint. To prevent entanglement, we introduce bias sinks. These are learnable tokens that indicate whether a training sample stems from a monocular or a multi-view dataset. During training, the model learns to produce incomplete 3D heads only when the monocular token is present. During inference, FlexAvatar then always uses the multi-view token for which the model has learned to produce complete 3D heads. This simple design allows to combine the generalizability from monocular data with the quality of multi-view data. FlexAvatar summary: - Input: Single-image, phone scan, or monocular video - Output: Full 360° head avatar - Expressive animations - Real-time rendering and animation - Generalization to any portrait - Create a new avatar in 2 minutes - Use bias sinks to combine 2D and 3D data 🏠 🌍 🎥 Great work by Tobias Kirschstein and Simon Giebenhain!

Matthias Niessner

95,991 просмотров • 7 месяцев назад

🔴 Finally! NVIDIA has finally made the code for Neuralangelo public! It has the ability to transform any video into a highly detailed 3D environment, and it's a technology related to but DIFFERENT from NeRF. 💡 Here's how it works: It takes a 2D video as input, showing an object, monument, building, landscape, etc., from various perspectives and analyzes details such as depth, size, and the shapes of objects. From this, the AI sketches an initial 3D model, similar to how an artist molds a figure. This representation is then refined to highlight more details, just as an artist would make the final touches when sculpting. The result is a 3D environment/model, perfect for use in any environment. Imagine the applications it will have for video games, cinema, virtual environments, VR, and more! 📽️🎮 💡 More details: A year ago, an article was presented on a groundbreaking technique called NVIDIA's Instant NeRF. This technique turns images into stunning 3D scenes in a short time, ideal for creating realistic models for video games and other applications. Although Instant NeRF had a lot of potential, the generated models were not perfect and often lacked detailed structures, appearing somewhat cartoonish. A year on, NVIDIA releases a new technique based on Instant NeRF, named Neuralangelo. This enhances the fidelity of surface structures. While NeRF reconstructs real objects in virtual environments from images or videos, Instant NeRF speeds up this process, and Neuralangelo further improves the quality, making the generated objects appear even more realistic when examined up close. Neuralangelo improves Instant NeRF's approach in two key ways related to the hash grid encoding technique: 1⃣ Numerical gradients have been used to compute higher-order derivatives as a smoothing operation. This optimizes the "hash grid" encoding using numerical rather than analytical gradients, providing a smoother input to the network that produces the 3D model. 2⃣ A "coarse-to-fine" optimization has been implemented in the hash grids to control different levels of detail. That is, they first focus on a smoothed version of the scene, and then refine it with more detailed updates. Well, as Arthur C. Clarke said, "Any sufficiently advanced technology is indistinguishable from magic."

Javi Lopez ⛩️

689,180 просмотров • 3 лет назад

Here's a quick test of a new tool called Krea. It's like the Real-time latent consistency model tests I've done with my photoshop and Dreams creations, but instead in a single application with faster refresh rates and higher quality output. I find that it strays too far from the drawing, and the editor is still quite limited but it's interesting to see nonetheless. It's not too difficult to imagine tech like this built right into Photoshop. And yes, I do know how to draw even if it doesn't show here 😅, I've been sculpting, painting and drawing for most of my life and I think it's exciting that tools like this will eventually help us break down some of the technical walls between our imagination and the visuals we create. And I'm not necessarily talking about drawing alone here because I don't see that going away as an art form, and I wouldn't want it to. But a hybrid of different art forms merging into a new, immersive art medium where anything is possible. Want to change the camera angle of your painting? Sure, go ahead. Want to make an animation of moving the camera around and seeing the light change in the scene as you do? In the example here, being able to move things around, change lighting and see it all reflected in the resulting image, is pretty amazing. Like a combination of the benefits of working in 3D, but with 2d tools. The moving version at the end of the video was made using Runways Gen-2. #art #ai

Martin Nebelong

1,668,469 просмотров • 2 лет назад

DALL-E 3 Double Exposure Images Using ChatGPT's Custom Instructions. 🔖 Bookmark and Repost! If you find this useful, please share it with others! You can now easily create double exposure images by using the syntax Color::subject1::subject2 You can either input this directly into DALL-E 3 before generating images or incorporate it into your custom instructions. This command will automatically use the fixed prompt to generate 4 different images, each with a unique seed value. Paste this under ChatGPT's Custom Instructions or DALL-3 before starting conversation. { DE": { "Instruction": "Using will only use generic prompt and update place holder varaibles and nothing else' Genric Prompt: "Construct a [Color] double exposure image where [Subject #1] is intricately superimposed within the confines of [Subject #2], all set against a stark white background.' Create 4 images with different seeds without modifying the generic prompt." You will first create two images. Within the same request, you will generate two more without asking for any input from the user. In general, you will always create 4 images. Your response should be something like this: 'Here are the first two images along with their seed details.' You must always provide the seed number details for that image after it's rendered. Then 'I'll generate the next two images.' And finally 'Here are the remaining two images along with their seed details.' You will always use wide aspect ratio and You must always provide the seed number. When command is activated display full instruction and also confirm that you will only use genric prompt and nothing else and update only varaiable", "CommandFormat": "color::subject1::subject2", "Response": { "Initial": "Creating images based on the provided subjects...", "AfterFirstSet": "Here are the first two images along with their prompt and seed details.", "AfterSecondSet": "Here are the remaining two images along with their prompt and seed details." }, "ActivationCommand": "/activate DE" } } Example: To activate type: /activate DE Notes Start giving prompt like this blue::mountain::wolf red::city skyline::dancer green::forest::lion If you found this post helpful, don't forget to hit the like and follow buttons, and share it with others who might find it useful.

AshutoshShrivastava

34,305 просмотров • 2 лет назад

A 19 YEAR OLD IN A ONE BEDROOM APARTMENT MAKES $15,200 A MONTH FROM CARTOONS SHE NEVER FILMED. YOUTUBE SENDS HER THE CHECK FOR EVERY VIEW. She does not draw. She does not animate. She does not own a camera. She opens one tab. YouTube. Types "kids shows." The top result pays its owner between $1,000 and $15,000 for every million views. She screenshots the number so you see it. Then she opens the second tab. An AI video generator. She pastes the transcript of a cartoon that already went viral. The tool builds a character from it. A toddler named BJ, blue onesie, biting a banana, rendered in bright kid-safe 3D. She never wrote BJ. She never voiced BJ. She copied a transcript, hit generate, and watched a cartoon she does not own get born in 90 seconds. Most creators argue about whether AI art is "real." She is not in that argument. She is in the tab where the ad revenue shows up. The "kids shows" search is not research. The "kids shows" search is the shelf she is about to stock. She posts the clip. YouTube runs ads on it. A toddler in Ohio watches BJ bite a banana and her balance moves by cents that compound into four figures. The cartoon is generated. The voice is synthesized. The ad check is real. She films herself in a black tube top explaining "how it works" because the explanation is the part that goes viral. The money is in the tab she keeps behind her. Save this post. The tool is an AI video generator. The niche is kids. The payout is YouTube's, and it does not ask who pressed generate.

ZER

2,465,841 просмотров • 23 дней назад