Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Liquid Playground / 05 Finally time to explore new tools. Quite fun this one. Simulated in JangaFX Software LiquiGen, rendered in Maxon with Maxon Redshift Sound design by @Manu_Sounds #liquigen #jangafx #cinema4d #redshift3d #simulation #3D

Dobromir Dyankov

3,211 subscribers

45,887 Aufrufe • vor 1 Jahr •via X (Twitter)

Bildung Kunst Wissenschaft & Technologie #liquigen #jangafx #cinema4d #redshift3d #simulation #3D

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Check out this gel-like simulation from the Liquid Playground series, created by Dobromir Dyankov using JangaFX Software's LiquiGen, Cinema 4D, and Redshift. See more: Maxon #liquigen #c4d #redshift #simulation

Check out this gel-like simulation from the Liquid Playground series, created by Dobromir Dyankov using JangaFX Software's LiquiGen, Cinema 4D, and Redshift. See more: Maxon #liquigen #c4d #redshift #simulation

80 LEVEL

30,346 Aufrufe • vor 1 Jahr

Real-time Fire and Liquid Simulation 🤯 Achieved using NVIDIA RTX PRO 6000 with 96 GB of VRAM 🤯🤯 Only Nerds like us get why this awesome!! Anybody who worked with 3D softwares knows how hard is it to learn and work with particle simulations in Real-time. Thanks to Dell and JangaFX Software we saw this Demo using the amazing softwares of EmberGen and LiquiGen where newcomers can quickly learn and execute complex simulations in real-time. Best part is you can export these simulations and bring it to your 3D software of choice such as Unreal Engine or Blender 🔶 😎😎

Real-time Fire and Liquid Simulation 🤯 Achieved using NVIDIA RTX PRO 6000 with 96 GB of VRAM 🤯🤯 Only Nerds like us get why this awesome!! Anybody who worked with 3D softwares knows how hard is it to learn and work with particle simulations in Real-time. Thanks to Dell and JangaFX Software we saw this Demo using the amazing softwares of EmberGen and LiquiGen where newcomers can quickly learn and execute complex simulations in real-time. Best part is you can export these simulations and bring it to your 3D software of choice such as Unreal Engine or Blender 🔶 😎😎

Bad Decisions Studio

39,942 Aufrufe • vor 11 Monaten

🚀 Bring your ideas to life with #MaxonOne! 🌱 Our Spring release is a creative powerhouse that makes anything possible. 🌟 We're thrilled to deliver powerful new features that enhance the creative potential of motion graphics artists, compositors, editors, and game artists, with updates across Maxon One: ✅ #Cinema4D ✅ #ZBrush ✅ #ZBrushForiPad ✅ #RedGiant ✅ #Redshift ✅ #MaxonStudio ✅ #Cineware This isn’t just software; it’s the pulse of a limitless creative community. 💓🌍🌎🌏

🚀 Bring your ideas to life with #MaxonOne! 🌱 Our Spring release is a creative powerhouse that makes anything possible. 🌟 We're thrilled to deliver powerful new features that enhance the creative potential of motion graphics artists, compositors, editors, and game artists, with updates across Maxon One: ✅ #Cinema4D ✅ #ZBrush ✅ #ZBrushForiPad ✅ #RedGiant ✅ #Redshift ✅ #MaxonStudio ✅ #Cineware This isn’t just software; it’s the pulse of a limitless creative community. 💓🌍🌎🌏

Maxon

13,153 Aufrufe • vor 1 Jahr

You as agent: super cool Gemini demo by Filip Havlena👤🍎🖥️ We've been experimenting with agentic interfaces at Creative Lab and this one was so fun: he turned himself into a 3d agent with spatial awareness + reasoning loop + tools (docs/emails with Google Workspace CLI). Made in Three.js. Real-time screencap ▶️

You as agent: super cool Gemini demo by Filip Havlena👤🍎🖥️ We've been experimenting with agentic interfaces at Creative Lab and this one was so fun: he turned himself into a 3d agent with spatial awareness + reasoning loop + tools (docs/emails with Google Workspace CLI). Made in Three.js. Real-time screencap ▶️

Alexander Chen

39,434 Aufrufe • vor 4 Monaten

NEW VIDEO! We wanted to cherish some of our most fun and beautiful moments in our first visit to México with our own rendition of 🌳FOREST INTERLUDE🌳 from Donkey Kong Country 2, originally composed by the one and only David Wise | Music Composer & Sound Design VGM, available in our latest album "CRITICAL HIT"

NEW VIDEO! We wanted to cherish some of our most fun and beautiful moments in our first visit to México with our own rendition of 🌳FOREST INTERLUDE🌳 from Donkey Kong Country 2, originally composed by the one and only David Wise | Music Composer & Sound Design VGM, available in our latest album "CRITICAL HIT"

Pokérus Project | now on bsky

10,732 Aufrufe • vor 1 Jahr

After testing every AI toolchain for a year, this is the only workflow I'm endorsing: Figma → Claude Code (Sonnet 4.5) Here's how I'm building Lorelight: 1. Design the happy path in Figma 2. Feed it to Claude Code via Figma MCP 3. Get a strong foundation in 15 minutes 4. Spend the next 50+ hours iterating (still design work) The design/dev loop that used to be isolated is finally closing. I even built a complete design system in within minutes. Custom components, living documentation, accessibility rules all from the Figma designs. If you're measuring the effectiveness of these tools by whether they can "one-shot" a design, you're consuming too many AI hype commercials. Log off. Then go actually build something with your hands. With AI. With your taste. Stop trying to "one-shot" good shit. And go make stuff with the new tools. For designers who want to ship real products, this is the stack that's getting it done.

After testing every AI toolchain for a year, this is the only workflow I'm endorsing: Figma → Claude Code (Sonnet 4.5) Here's how I'm building Lorelight: 1. Design the happy path in Figma 2. Feed it to Claude Code via Figma MCP 3. Get a strong foundation in 15 minutes 4. Spend the next 50+ hours iterating (still design work) The design/dev loop that used to be isolated is finally closing. I even built a complete design system in within minutes. Custom components, living documentation, accessibility rules all from the Figma designs. If you're measuring the effectiveness of these tools by whether they can "one-shot" a design, you're consuming too many AI hype commercials. Log off. Then go actually build something with your hands. With AI. With your taste. Stop trying to "one-shot" good shit. And go make stuff with the new tools. For designers who want to ship real products, this is the stack that's getting it done.

Tommy Geoco

22,905 Aufrufe • vor 9 Monaten

Created a flag on the moon based with hyper-realistic physics with v0: The flag is simulated as a grid of particles connected by springs (cloth simulation). It's influenced by gravity, wind, air resistance. The initial approach v0 one-shotted was quite simplistic, but giving it references to more sophisticated approaches in C++ and Rust quickly improved it. It was mind-blowing it to see it deeply understand the physics and port new ideas across languages. It simulated a real sun for lighting, and procedurally generated moon-like terrain with craters. I found a sick free texture on Reddit that I uploaded and it layered on top. Another epic moment was watching it improve performance. It seamlessly moved the physics simulation to a Web Worker. (Kinda cool to see given we just improved support for them in Turbopack 😉)

Created a flag on the moon based with hyper-realistic physics with v0: The flag is simulated as a grid of particles connected by springs (cloth simulation). It's influenced by gravity, wind, air resistance. The initial approach v0 one-shotted was quite simplistic, but giving it references to more sophisticated approaches in C++ and Rust quickly improved it. It was mind-blowing it to see it deeply understand the physics and port new ideas across languages. It simulated a real sun for lighting, and procedurally generated moon-like terrain with craters. I found a sick free texture on Reddit that I uploaded and it layered on top. Another epic moment was watching it improve performance. It seamlessly moved the physics simulation to a Web Worker. (Kinda cool to see given we just improved support for them in Turbopack 😉)

Guillermo Rauch

534,602 Aufrufe • vor 3 Monaten

Water simulation has always been the holy grail of FX, requiring an incredible amount of computation to get right. For Battleship (2012), ILM had to reinvent their water systems to allow them to be able to deliver the show with it's plethora of shots in water. The alien ships are hundreds of feet long and therefore displace massive amounts of water. This, coupled with them making aggressive movements meant that hundreds of millions of droplets of water had to be simulated per shot to get a believable look of water interaction. Before their new water systems, artists were only able to get "1-2 simulations done per day", but with the new system they were able to "do many simulations per day", increasing the iterations and therefore the creative freedom and expression for the artists. ILM has since reworked their water systems multiple times, now being more powerful and accurate than ever, utilizing more powerful hardware and more accurate and efficient simulation algorithms. Today, water simulation is something that can be done a lot easier than before, even being able to do it in real-time with tools such as Unreal Engine and LiquiGen. Houdini is currently the most powerful tool for simulating water or any FX at very high fidelities. Check out some FX flipbooks from ILMs VFX breakdown showing their water work below:

Water simulation has always been the holy grail of FX, requiring an incredible amount of computation to get right. For Battleship (2012), ILM had to reinvent their water systems to allow them to be able to deliver the show with it's plethora of shots in water. The alien ships are hundreds of feet long and therefore displace massive amounts of water. This, coupled with them making aggressive movements meant that hundreds of millions of droplets of water had to be simulated per shot to get a believable look of water interaction. Before their new water systems, artists were only able to get "1-2 simulations done per day", but with the new system they were able to "do many simulations per day", increasing the iterations and therefore the creative freedom and expression for the artists. ILM has since reworked their water systems multiple times, now being more powerful and accurate than ever, utilizing more powerful hardware and more accurate and efficient simulation algorithms. Today, water simulation is something that can be done a lot easier than before, even being able to do it in real-time with tools such as Unreal Engine and LiquiGen. Houdini is currently the most powerful tool for simulating water or any FX at very high fidelities. Check out some FX flipbooks from ILMs VFX breakdown showing their water work below:

Rassoul Edji

86,738 Aufrufe • vor 1 Jahr

Finally found the time to promote this thing properly. Also gave it a name... MECHAWINGS!!! 😁 A mechanical wing rig for blender with just ONE easy controller. You can get it in my new Patreon shop! Or by subscribing 😉 🔗 Link in my Bio 🔗 #b3d

Finally found the time to promote this thing properly. Also gave it a name... MECHAWINGS!!! 😁 A mechanical wing rig for blender with just ONE easy controller. You can get it in my new Patreon shop! Or by subscribing 😉 🔗 Link in my Bio 🔗 #b3d

DemNikoArt 🔶

16,113 Aufrufe • vor 1 Jahr

In my AI filming career, there is a niche I really enjoy working with and where AI Video brings lots of value. That niche is Real Estate. AI is perfect to animate 3D renders and to create b-role for mood. Here I will try to answer two questions: 1 - How to make a real estate video? 2- How to sell a house? This was a work done based on 3D Renders (provided by The Creative Farm, more on them below). Sound design made by noon (feel free to DM her for great sound design) Below I will break down a bit of the process, the prompts, timespans, the tools I used as well as some secrets this time around (feeling pretty today) It will be a bit long, so bookmark it or save it for later if you have the interest but lack the time.

In my AI filming career, there is a niche I really enjoy working with and where AI Video brings lots of value. That niche is Real Estate. AI is perfect to animate 3D renders and to create b-role for mood. Here I will try to answer two questions: 1 - How to make a real estate video? 2- How to sell a house? This was a work done based on 3D Renders (provided by The Creative Farm, more on them below). Sound design made by noon (feel free to DM her for great sound design) Below I will break down a bit of the process, the prompts, timespans, the tools I used as well as some secrets this time around (feeling pretty today) It will be a bit long, so bookmark it or save it for later if you have the interest but lack the time.

FERA

218,120 Aufrufe • vor 11 Monaten

Before Half-Life... there was Fade to Black (1995), a truly brilliant action-adventure game developed by Delphine Software as a sequel to Flashback. It follows hero Conrad Hart, awakened from cryogenic sleep in an alien prison on the Moon. You have to explore, fight, solve puzzles, and navigate a sci-fi conspiracy in a fully 3D polygonal world. Back in the day, it felt revolutionary with smooth 3D graphics, dynamic camera angles, and a perfect blend of shooting, platforming, and adventure way before Tomb Raider took the crown. Story, atmosphere, pace, graphics, sound effects - it was the whole package. In 1995, this was a real masterclass.

Before Half-Life... there was Fade to Black (1995), a truly brilliant action-adventure game developed by Delphine Software as a sequel to Flashback. It follows hero Conrad Hart, awakened from cryogenic sleep in an alien prison on the Moon. You have to explore, fight, solve puzzles, and navigate a sci-fi conspiracy in a fully 3D polygonal world. Back in the day, it felt revolutionary with smooth 3D graphics, dynamic camera angles, and a perfect blend of shooting, platforming, and adventure way before Tomb Raider took the crown. Story, atmosphere, pace, graphics, sound effects - it was the whole package. In 1995, this was a real masterclass.

exQUIZitely 🕹️

131,059 Aufrufe • vor 2 Monaten

I’m not a 3D artist, but I had so much fun today testing Slop Unbound and creating cute creatures and models for the first time It’s FREE, experimental, easy to play with, and honestly, one of the most fun ways I’ve used AI for visual creation because you get real control instead of just prompting and hoping for the best You do not need to rely only on Blender anymore AI can simplify your workflow and make 3D creation feel much more accessible If you are into 3D art, character design, or just curious about new creative tools, you should absolutely test it. It’s fast, fun, and surprisingly addictive I’ll share more in the coming days, but for now, go try it yourself for FREE and share your results. I think you’re going to love it, link in the first comment 👇🏻

I’m not a 3D artist, but I had so much fun today testing Slop Unbound and creating cute creatures and models for the first time It’s FREE, experimental, easy to play with, and honestly, one of the most fun ways I’ve used AI for visual creation because you get real control instead of just prompting and hoping for the best You do not need to rely only on Blender anymore AI can simplify your workflow and make 3D creation feel much more accessible If you are into 3D art, character design, or just curious about new creative tools, you should absolutely test it. It’s fast, fun, and surprisingly addictive I’ll share more in the coming days, but for now, go try it yourself for FREE and share your results. I think you’re going to love it, link in the first comment 👇🏻

Amira Zairi

32,890 Aufrufe • vor 3 Monaten

Jensen Huang just said this is the greatest era in history to build software. AI agents will not kill software. They will do the exact opposite: create a massive new wave of software demand. at NVIDIA GTC Taipei 2026 "Click and type. We now replace that with explaining to the AI what we want, our intent, and the AI generates the code or uses tools to produce the necessary output. This is how computers are going to work in the future. This is Agentic AI. For two years, we've been building toward this, and now it has arrived. One of the big breakthroughs, of course, is tool use. A lot of people have said, “Jensen, AI is coming. Agentic AI is coming. Therefore, all the software companies are going to go out of business.” This is exactly the opposite. Because there are going to be so many agents, the world is no longer limited by the number of people. Therefore, those agents are going to use more tools than ever. This is actually an incredible time to be a software company. But the software has to be presented to the agent in a way that the agent can use it. This is a big breakthrough. And in fact, what we have done, as you know, what Nvidia’s treasure..." ---- From 'NVIDIA' YT channel (link in comment)

Jensen Huang just said this is the greatest era in history to build software. AI agents will not kill software. They will do the exact opposite: create a massive new wave of software demand. at NVIDIA GTC Taipei 2026 "Click and type. We now replace that with explaining to the AI what we want, our intent, and the AI generates the code or uses tools to produce the necessary output. This is how computers are going to work in the future. This is Agentic AI. For two years, we've been building toward this, and now it has arrived. One of the big breakthroughs, of course, is tool use. A lot of people have said, “Jensen, AI is coming. Agentic AI is coming. Therefore, all the software companies are going to go out of business.” This is exactly the opposite. Because there are going to be so many agents, the world is no longer limited by the number of people. Therefore, those agents are going to use more tools than ever. This is actually an incredible time to be a software company. But the software has to be presented to the agent in a way that the agent can use it. This is a big breakthrough. And in fact, what we have done, as you know, what Nvidia’s treasure..." ---- From 'NVIDIA' YT channel (link in comment)

Rohan Paul

16,584 Aufrufe • vor 1 Monat

🌊 GTA 6 will take liquid simulation to a whole new level. Rockstar is reportedly using fully dynamic fluid physics, calculated in real time — not just animations or pre-baked effects like in most open-world games. That means: • Water will react naturally to wind, waves and storms • Liquids will flow, splash and deform based on collisions • Objects falling into water will displace it physically • Puddles, floods and wet surfaces will evolve dynamically • Drinks, blood, gasoline and more will move like real liquids All of this inside a massive, open, living world. 🤯🔥 This tech could drastically change gameplay in a tropical map like Miami / Vice City — storms, hurricanes, beaches, boats… everything becomes more immersive and unpredictable. And just imagine the details indoors: Glasses pouring realistically, liquid containers that actually react to motion, bars feeling truly alive… 🍻 To have some fun with the idea, I tried creating a GTA 6-style beer using 3D tools. Let’s just say… it’s far from Rockstar level 😅 I have zero experience in fluid simulation or 3D animation, so please be kind! 📸 Fan-made concept only — nothing here is official. How crazy do you think Rockstar’s fluid system will be? 👀🌴 #GTA6 #GTAVI #RockstarGames #NextGenPhysics #FluidSimulation #GameTech #ViceCity

🌊 GTA 6 will take liquid simulation to a whole new level. Rockstar is reportedly using fully dynamic fluid physics, calculated in real time — not just animations or pre-baked effects like in most open-world games. That means: • Water will react naturally to wind, waves and storms • Liquids will flow, splash and deform based on collisions • Objects falling into water will displace it physically • Puddles, floods and wet surfaces will evolve dynamically • Drinks, blood, gasoline and more will move like real liquids All of this inside a massive, open, living world. 🤯🔥 This tech could drastically change gameplay in a tropical map like Miami / Vice City — storms, hurricanes, beaches, boats… everything becomes more immersive and unpredictable. And just imagine the details indoors: Glasses pouring realistically, liquid containers that actually react to motion, bars feeling truly alive… 🍻 To have some fun with the idea, I tried creating a GTA 6-style beer using 3D tools. Let’s just say… it’s far from Rockstar level 😅 I have zero experience in fluid simulation or 3D animation, so please be kind! 📸 Fan-made concept only — nothing here is official. How crazy do you think Rockstar’s fluid system will be? 👀🌴 #GTA6 #GTAVI #RockstarGames #NextGenPhysics #FluidSimulation #GameTech #ViceCity

Zap Actu GTA6

122,308 Aufrufe • vor 7 Monaten

Claude Fable 5 is insane again. The first time, I only used it to build a 3D game clone. I just used Fable 5 inside Rork to build a new tower defense game in about 1 hour. From the initial idea → level design → heroes & monsters → in-game economy → game art → sound effects → code Each level will have different difficulty, different map upgrades, different monsters, and different bosses. Each tower weapon has unique skills, counters other weapons in different ways, and the heroes are controllable with skills that can support each other... All in one flow. This is getting crazy me again. 🙃

Claude Fable 5 is insane again. The first time, I only used it to build a 3D game clone. I just used Fable 5 inside Rork to build a new tower defense game in about 1 hour. From the initial idea → level design → heroes & monsters → in-game economy → game art → sound effects → code Each level will have different difficulty, different map upgrades, different monsters, and different bosses. Each tower weapon has unique skills, counters other weapons in different ways, and the heroes are controllable with skills that can support each other... All in one flow. This is getting crazy me again. 🙃

Alex Nguyen

99,400 Aufrufe • vor 21 Tagen

This is the power of VFX and AI now.. Creating volumetric effects in AI and then using AfterEffects and Premiere to layer and edit + Topaz Labs to upscale, you can visually achieve the same effects that would have taken me A LOT longer in Cinema4D/Blender. This is really amazing, only because if I can do this now with what we have, imagine where it will be in 6-mo 🔥 it will only get better. Soon we’ll be able to do a lot more within 3D environments with gaussian splats and volumetric effects. It won’t be long after that before Cinema, blender and others integrate gaussian splats and generative prompting directly into the software design and render process. At that point things are going to get REALLY BADASS! I’m chillin tonight, finally a night off lol. But tomorrow I’ll show you guys how I did this ❤️

This is the power of VFX and AI now.. Creating volumetric effects in AI and then using AfterEffects and Premiere to layer and edit + Topaz Labs to upscale, you can visually achieve the same effects that would have taken me A LOT longer in Cinema4D/Blender. This is really amazing, only because if I can do this now with what we have, imagine where it will be in 6-mo 🔥 it will only get better. Soon we’ll be able to do a lot more within 3D environments with gaussian splats and volumetric effects. It won’t be long after that before Cinema, blender and others integrate gaussian splats and generative prompting directly into the software design and render process. At that point things are going to get REALLY BADASS! I’m chillin tonight, finally a night off lol. But tomorrow I’ll show you guys how I did this ❤️

Dustin Hollywood

19,760 Aufrufe • vor 8 Monaten

Even the most successful websites eventually need a redesign. Take Stripe for example. After six years with the same homepage, Stripe recently unveiled a brand new site that reflects how the company has evolved over the past few years. So when is the time right for a new landing page? And what should you prioritize in the redesign? In this episode of Design Review, YC’s Aaron Epstein sat down Stripe’s Head of Design Katie Dill (Katie Dill) to pull the curtain back on their high profile redesign and to discuss how their team is evolving in a world dominated by new AI design tools. 00:30 - The Old Site 05:30 - “What is the point of a website?” 06:00 - Breaking down the new homepage 07:10 - The GDP counter (and how Stripe builds trust) 08:45 - The Bento Box that fixed the homepage 11:37 - Why great design feels effortless 15:40 - Craft, iteration, and getting it right 19:11 - Inside Stripe’s design process 27:10 - How AI is changing design 35:00 - The fight against mediocrity 40:00 - Walking the Store: Stripe’s Secret to Better Product Design

Y Combinator

86,552 Aufrufe • vor 3 Monaten

CapCut Design Studio changed the way I think about creating with AI. For a long time, creating with AI felt less like creating and more like managing files—generate, export, switch tools, edit, re-upload, repeat. The ideas were flowing, but the workflow kept getting in the way. What's changing now isn't just image quality, it's the creative process itself. This video shows how I used CapCut Design Studio to build an autumn fashion poster from scratch. I started on the infinite canvas to explore styles, then used the pencil to easily remove unwanted elements and even swap out outfits to find the perfect look for my model. Also, I used the text editing feature to instantly translate my poster into multiple languages without losing the original aesthetic. If I ever need more options, the 'One More' feature automatically creates multiple creative variations with one click. Instead of switching between multiple tools, I was able to keep the entire creative process connected in one place. AI design is moving from "generate and download" to "create, refine, and scale"—and that's the shift I'm most excited about. What's the biggest bottleneck in your current design workflow? #capcutdesignstudio #capcut

CapCut Design Studio changed the way I think about creating with AI. For a long time, creating with AI felt less like creating and more like managing files—generate, export, switch tools, edit, re-upload, repeat. The ideas were flowing, but the workflow kept getting in the way. What's changing now isn't just image quality, it's the creative process itself. This video shows how I used CapCut Design Studio to build an autumn fashion poster from scratch. I started on the infinite canvas to explore styles, then used the pencil to easily remove unwanted elements and even swap out outfits to find the perfect look for my model. Also, I used the text editing feature to instantly translate my poster into multiple languages without losing the original aesthetic. If I ever need more options, the 'One More' feature automatically creates multiple creative variations with one click. Instead of switching between multiple tools, I was able to keep the entire creative process connected in one place. AI design is moving from "generate and download" to "create, refine, and scale"—and that's the shift I'm most excited about. What's the biggest bottleneck in your current design workflow? #capcutdesignstudio #capcut

Farlane

41,971 Aufrufe • vor 1 Monat

Exciting updates on Project GR00T! We discover a systematic way to scale up robot data, tackling the most painful pain point in robotics. The idea is simple: human collects demonstration on a real robot, and we multiply that data 1000x or more in simulation. Let’s break it down: 1. We use Apple Vision Pro (yes!!) to give the human operator first person control of the humanoid. Vision Pro parses human hand pose and retargets the motion to the robot hand, all in real time. From the human’s point of view, they are immersed in another body like the Avatar. Teleoperation is slow and time-consuming, but we can afford to collect a small amount of data. 2. We use RoboCasa, a generative simulation framework, to multiply the demonstration data by varying the visual appearance and layout of the environment. In Jensen’s keynote video below, the humanoid is now placing the cup in hundreds of kitchens with a huge diversity of textures, furniture, and object placement. We only have 1 physical kitchen at the GEAR Lab in NVIDIA HQ, but we can conjure up infinite ones in simulation. 3. Finally, we apply MimicGen, a technique to multiply the above data even more by varying the motion of the robot. MimicGen generates vast number of new action trajectories based on the original human data, and filters out failed ones (e.g. those that drop the cup) to form a much larger dataset. To sum up, given 1 human trajectory with Vision Pro -> RoboCasa produces N (varying visuals) -> MimicGen further augments to NxM (varying motions). This is the way to trade compute for expensive human data by GPU-accelerated simulation. A while ago, I mentioned that teleoperation is fundamentally not scalable, because we are always limited by 24 hrs/robot/day in the world of atoms. Our new GR00T synthetic data pipeline breaks this barrier in the world of bits. Scaling has been so much fun for LLMs, and it's finally our turn to have fun in robotics! We are building tools to enable everyone in the ecosystem to scale up with us. Links in thread:

Jim Fan

364,380 Aufrufe • vor 2 Jahren