Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

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 🧵👇

311,229 görüntüleme • 1 yıl önce •via X (Twitter)

24 Yorum

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

Available NOW at Magnific 🪄 for all users! 👇

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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!

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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%

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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%

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

👉 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.

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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 🤯

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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).

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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%

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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 :)

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

"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"

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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%

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

You have a TON more of examples here:

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

🔴 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.

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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%

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

Show me your best results!!!

Javi Lopez ⛩️ profil fotoğrafı
Javi Lopez ⛩️1 yıl önce

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!

AndaSeat profil fotoğrafı
AndaSeat1 yıl önce

✨ 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

Benzer Videolar

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

96,371 görüntüleme • 9 ay önce

👇7 simple steps using ChatGPT + Google Flow to create AI animated videos? 🤖🎬 Want the complete AI Video Creation Blueprint eBook? with step-by-step instructions + ready-to-use prompts. 🔥Comment “AI” and I’ll send you the link. ✅ Step 1: Choose a Character Choose your main character. Example: a cute superhero monkey with a red cape. ✅ Step 2: Get 10 Story Ideas Ask ChatGPT to generate 10 short, emotional and engaging story ideas around your character. ✅ Step 3: Create the Complete Story Select your favorite idea and ask ChatGPT to turn it into a complete 30–40 second story with a strong hook, emotion and ending. ✅ Step 4: Create a Reference Image Use ChatGPT to create your main reference image. This helps maintain the same character look throughout the video. ✅ Step 5: Create Scene-by-Scene Prompts Ask ChatGPT to divide your story into scenes and create detailed Google Flow prompts for every scene. ✅ Step 6: Create Clips in Google Flow Upload your reference image, use the prompts and generate each scene as a separate video clip. ✅ Step 7: Edit Your Video Combine all the clips using any video editing software. Add music, sound effects and text—and your AI video is ready! 🎥 Disclaimer: AI-generated content can be monetized, provided it follows the respective platform’s monetization, originality, copyright, and AI content guidelines. #SriramBenur #AIVideo #AIContentCreation See less

Mohini Maheshwari

26,025 görüntüleme • 8 gün önce

🔴 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 ⛩️

689,325 görüntüleme • 3 yıl önce

Probably I vibe coded a lil startup here? 😭 It has been such a loooong wish of mine to build some kind of 3D experience where I could customize a T-shirt in literally any way possible and it’s finally here. Built with Three.js using GPT 6 Astra, this is a full 3D T-shirt customization studio where you can visualize and customize a realistic shirt directly in the browser. That T-shirt itself was modeled by Astra using Tripo right inside Codex through Tripo Plugin. And you can pretty much do anything with it. You can paint directly on any side of the shirt using different brushes and colors, or even spray paint it in real time just like you would spray on a wall. There are also stickers generated using GPT Image 2.5 that you can place anywhere on the T-shirt, resize, reposition, recolor, layer, and customize however you want. On top of that, there’s support for things like fabric customization, sizing, layers, colors, and even wind simulation to push the realism a little further. Once you’re done, you can export the entire design as a 3D view or export individual images of the T-shirt so you could technically take the design, print it, and maybe even sell it. One of my favorite parts is how the spray painting effect works directly on the 3D T-shirt in real time. I also loved seeing how Astra managed to keep the whole experience performant across devices using a custom BVH implementation along with several CPU side optimizations. And the process of building it was super simple. I generated the initial studio design using GPT Image 2.5, gave that image to Astra with the Tripo Plugin enabled, and it basically handled everything from there. I didn’t have to separately generate a 3D reference, upload it to Tripo, download the asset, give it back to Astra, or manually coordinate any of that. Astra handled the entire flow on its own without needing any additional input from me. Really happy with both the process and how the final result turned out. Live:

The Bugged Dev

51,942 görüntüleme • 20 gün önce

Trained a humanoid entirely in a 3D scan of the office. Zero real-world fine-tuning. It just walked in and worked. RL needs hundreds of thousands of attempts, and real robots can't afford to crash. A misjudged gap or a glass door collision breaks hardware and costs hours resetting. So you train in a sim. But sim policies usually train on randomized, untextured geometry; depth is easy to fake. The robot learns structure, not the real world: no materials, no lighting, no idea what anything actually is. RGB cameras carry all of that but training RGB policies in generic fake worlds won’t generalize to the real world. Niantic Spatial 🌎 Scaniverse reconstructs your scan of the real deployment site. One 360° camera walkthrough → photorealistic 3D Gaussian splat at metric scale → collision mesh pulled from the same reconstruction, so vision and physics match exactly. Drops straight into NVIDIA Isaac Sim/Lab, no manual conversion. Flexion simulation-first approach then seamlessly enables the training of RGB-only nav policies inside that reconstruction. With added domain randomization + large image encoders for robustness, this deploys straight to hardware. No real-world fine-tuning. Deployment: months of on-site adaptation → days. Tune into the NVIDIA livestream on 12 August to hear how these companies are closing the sim2real gap: NVIDIA Robotics ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

119,739 görüntüleme • 1 ay önce