Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

3D Artist Tawan Sunflower has released a step-by-step narrated YouTube tutorial on creating a simple but effective stylized real-time rain and splashes simulation with Blender 🔶's Geometry Nodes. Get free project files here: #b3d #blender #geonodes

50,948 Aufrufe • vor 2 Jahren •via X (Twitter)

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

It is spectacularly easy to get shadowbanned on X One wrong reply and your reputation can go up in flames Luckily there are a few steps you can take to reduce your odds of getting banned by 10x 🔶First step: Don't reply from notifications Notifications are a minefield because hidden, spammy replies are shown to you as normal replies It's never worth replying from here. You risk replying to low reputation accounts without even knowing it Just reply normally from the posts so you know what's marked as spam 🔶Second step: turn off sensitive content in your settings Engaging with sensitive content could hurt your reputation score Little do you know, but a ton of content that’s not actually sensitive, gets marked as sensitive Turn off sensitive content in your settings, then become shocked when you see how much innocent content is marked as sensitive (tutorial video below) 🔶Third Step: Check for following/follower ratio before hitting reply Easy check to take before every reply This is brought up in the code 100 different times as a reputation destroyer Make sure to hover over people’s name before replying and check for their ratio 60% followers to following is the ratio cut off point 🔶Fourth Step: Don’t feed the trolls There are 100+ labels for toxic behavior in the code Engaging with anyone with these labels hurts your reputation For some of these labels they can be applied just for “insults” Best not to engage with any trolls (also good for your mental health) 🔶Fifth Step: Get the blue checkmark The easiest step of them all It’s coded into the algo that if you have the blue checkmark you can’t be marked as spam Yes it costs money, but if you have any serious interest of growing on this platform, it’s a must have 🔶🔶🔶🔶🔶 I've said this before, but it absolutely stinks you have to take all these steps just to get some reach At times, it can be like walking in a minefield It doesn't exactly make a social media platform feel very 'social' I have high confidence Elon is working to fix this. He's mentioned it before. But in the meantime, take these steps and be careful. It's worth it if you care about being seen on this platform. I'm 100% confident brighter days are ahead. Let me know below if any of these tips helped and what other steps you take to avoid getting banned.

Alex Finn

300,049 Aufrufe • vor 2 Jahren

📢 Our lab has been exploring 3D world models for years — and we’re thrilled to share **PhysTwin**: a milestone that reconstructs object appearance, geometry, and dynamics from just a few seconds of interaction! Led by the amazing Hanxiao Jiang 👉 PhysTwin combines **Gaussian splatting** with **inverse dynamics optimization** based on simple **spring-mass** systems. ⚙️ The result? Real-time, action-conditioned 3D video prediction under novel interactions (i.e., 3D world models). 🔑 A few key takeaways: 1. Having the right structure (e.g., particles/masses) helps navigate the trade-off between sample efficiency, generalization, and broad applicability. 2. Visual foundation models (VFMs) have matured to the point where they can provide rich supervision for world modeling (e.g., tracking, shape completion). 3. Beyond VFMs, many crucial components have come together in recent years: Gaussian splats for rendering, NVIDIA Warp for high-performance simulation, and scene/asset generation from a wide range of labs and companies. The future of 3D world models is looking bright! ✨ 4. The resulting digital twin supports a wide range of downstream applications—especially in data generation and policy evaluation, thanks to its realistic rendering and simulation capabilities. 🎥 All code and data to reproduce the results, along with interactive demos, are available on the website. Check the following visualizations of: (1) observations, (2) reconstructed state/actions, (3) interactive digital twins, and (4) the overlays between real-world robot teleoperation and our model’s open-loop predictions.

Yunzhu Li

25,279 Aufrufe • vor 1 Jahr

Trained on zero real-world data. Learned to walk, pick up boxes, and follow multi-step instructions... in the REAL world. ( 📌 Paper below) Researchers from Amazon FAR, Berkeley, Stanford, and CMU scanned real rooms with an iPhone, rebuilt them as 3D Gaussian Splatting scenes, then generated 48,000 synthetic trajectories of a Unitree G1 walking, grasping, and placing objects inside those virtual replicas. They rendered the robot's first-person camera view from each run and paired it with the matching language instruction and motion data. That's the dataset every humanoid team needs and nobody has: synced egocentric video + language + kinematics, at scale. Instead of collecting it in the real world, they manufactured it. They trained a vision-language-kinematics policy on that synthetic data alone, then deployed it on the physical G1 across five task types: navigation to a named object, lifting boxes of three different sizes with no per-size tuning, chained multi-step tasks, robustness to mid-task layout changes and flickering lights, and multi-minute long-horizon runs. No real-world fine-tuning at any point. Real-world interaction data has been the hard limit on humanoid learning... slow, expensive, and small. If scanning a room once and synthesizing thousands of labeled interactions holds up as a general recipe, that limit moves. Data stops being the bottleneck robotics teams have to solve for. 📌 Paper: Project: ——- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

12,950 Aufrufe • vor 9 Tagen

Claude Cowork is f*cking ridiculous 🤯 One prompt → competitive research, creative briefs, 15 hook variations, and a full performance dashboard. All saved as real files on your computer. All inside Claude Desktop. If you're spending hours every week copy-pasting between tools, pulling competitor ads manually, writing briefs from scratch, and building reports in spreadsheets ... Claude Cowork eliminates the entire loop: → Point it at your project folder with brand voice + context files → It asks YOU clarifying questions instead of guessing → It builds a multi-step plan and executes while you step away → It creates real .docx, .xlsx, .pptx files — not chat responses → It connects to Slack, Google Drive, Airtable, and 50+ tools live No copy-pasting between tools. No babysitting the AI mid-task. No downloading and re-uploading files. What you get: → Competitive research synthesized into actionable creative angles → Ad briefs, hooks, and scripts generated in your brand voice → Interactive HTML dashboards built from your own customer data → Weekly performance reports created while you're getting coffee Built 100% inside Claude Desktop with skills, plugins, and connectors. I put together a full DTC playbook: 10 workflows with copy-paste prompts, the exact setup process, and the weekly operating rhythm I use. Want it for free? > Like this post > Comment "COWORK" And I'll send it over (must be following so I can DM)

Utkarsh Sharma

29,875 Aufrufe • vor 4 Monaten