Jonathan Stephens
@jonstephens85 • 13,705 subscribers
Synthetic Data | World Models | Radiance Fields | Computer Vision | GenAI | Chief Evangelist at @LightwheelAI
Shorts
Videos
The most impressive robot demo I saw at IROS was also one of the simplest. Agility had Digit autonomously picking objects from a tote, walking them across the booth, and repeating the task until the tote was empty. I talked with the mastermind behind the demo, Chris Paxton, about how the demo worked: ~500 real-world teleoperated demonstrations for the manipulation policy A whole-body controller trained in simulation Deployed into a completely unseen environment at IROS Roughly 80–90% success in the booth despite crowds, harsh lighting, and plenty of distractions Chris was also very clear that this is still research, not something currently running in production. It was a glimpse at how quickly learned manipulation policies can move from controlled training environments into the messy real world. Lukas Ziegler you might find this convo interesting. #IROS2026 #AgilityRobotics #PhysicalAI #RobotLearning
Jonathan Stephens26,338 Aufrufe • vor 16 Tagen
This is by far the best 3DGS scan of my kitchen. I love how the light reflects on my dining room table from the bay windows. I still struggle to be able to deal with the high dynamic range between the inside and outside though. The windows are overexposed which means you can't see anything outside well. Does anyone have a good solution to this? Shoot raw and globally adjust? I would love to share notes with anyone working on this challenge at #CVPR2026 #3DGS #ComputerVision
Jonathan Stephens22,254 Aufrufe • vor 4 Monaten
SimReady assets generated from a simple prompt! No 3D artist needed. Last week at SIGGRAPH, Lightwheel launched SimReadyGen - an agentic SimReady asset generator. Since then, our team has been improving the experience, including an in browser asset viewer. I recorded this quick video to give you a peek of what the generator looks like. Want to give it a try? You an access it today at #PhysicalAI #Simulation
Jonathan Stephens11,583 Aufrufe • vor 2 Monaten
Do you think you could train a robot to fold clothes better than any other robotics team? Now is your chance to prove it...and win actual cash prizes! The 2026 LeHome Challenge is live! Here is the details: Phase 1: Train your best clothes folding policy in simulation with 4 garment types, each presenting unique manipulation challenges. The types of garments include long-sleeved tops, short-sleeved tops, long pants, and shorts. The simulation environment uses Lightwheel's physics accurate deformable assets and built on NVIDIA Omniverse's Isaac Sim/Lab. Submit your code by April 5th for a chance to win up to $2,000 and a SO-101 Arm Kit. Phase 2: The top 8 teams will compete in person at ICRA on June 4th to see who's policy performs best with real clothes. The top place team will take home $20,000! Learn more and sign up for the LeHome Challenge here: Robert Scoble Chris Paxton The Humanoid Hub Harrison Kinsley #Robotics
Jonathan Stephens24,003 Aufrufe • vor 8 Monaten
The team at NVIDIA AI Developer really nailed it with ViPE!! This SLAM system generated pose, depth, point clouds, etc, in a few seconds! Now to get the output to work with GEN3C... If you are interested in running this yourself, it took about 5 minutes to get up and running from scratch.
Jonathan Stephens34,947 Aufrufe • vor 1 Jahr
This is the most Disney thing I have ever seen when it comes to robotics! They trained robots to fall whimsically and in style. Why? Because not only does it protect its components, it looks cool! Watch the full video here: #Robotics #ReinforcementLearning #PhysicalAI
Jonathan Stephens24,707 Aufrufe • vor 11 Monaten
I created a 3D Gaussian Splat of my kitchen using 1/3rd the images it used to take me by using NVIDIA AI Developer's 3DGRUT! I can now use 180 degree fisheye images and ray tracing to make detailed splats. The only reason the scene isn't sharper is because my input images weren't super sharp - when I took the images back in October, I was still learning to use the lens. I plan to make a "first reactions/overview video". Tutorial after that. For reference, this took 206 images and the ultrawide on my iPhone took 608 images to capture. #3D #AEC #Computervision
Jonathan Stephens30,717 Aufrufe • vor 1 Jahr
The greatest Gaussian Splatting VR demo ever? Captured live at NVIDIA GTC — Mariya Sha experiencing 4D Gaussian Splatting in VR for the very first time. You can tell just from her reaction that this was something else! The scene was optimized using QUEEN, NVIDIA AI Developer's advanced research project for fast radiance field rendering. It’s an incredible look at what happens when high fidelity static and dynamic gaussian splats meets VR. Watch the full video on YouTube: Robert Scoble Did you get to do this one? #VR #3D #VFX
Jonathan Stephens28,114 Aufrufe • vor 1 Jahr
![Video world models look great but often ignore physics and 3D geometry. New research achieves a 64% reduction in navigation error by teaching AI to prioritize physical reality over visual "eye-candy". Here’s how they did it: #GenerativeAI #Robotics [1/n]](https://image.24vids.com/tw-2010824003968250220/amplify_video_thumb/2010822641788014592/img/xmrhIqj1s-2JL7T-.jpg)
Video world models look great but often ignore physics and 3D geometry. New research achieves a 64% reduction in navigation error by teaching AI to prioritize physical reality over visual "eye-candy". Here’s how they did it: #GenerativeAI #Robotics [1/n]
Jonathan Stephens15,674 Aufrufe • vor 9 Monaten
I’ve rebuilt my kitchen with photogrammetry, gaussian splats, LiDAR, and just about any other method you could think of...so I had to see what World Labs' Marble could do with only 8 photos. Here’s a first look at what worked, and what didn’t. Link to full video: #3D #GenerativeAI #AEC #AI
Jonathan Stephens16,501 Aufrufe • vor 11 Monaten
Big news! World Labs' Marble is officially compatible with NVIDIA Robotics' Isaac Sim. Instead of showcasing their highlight reel, our team at Lightwheel created a full environment from a 360 pano using Marble and showed you can train directly in sim. The only non-generated assets are the robot arms, coffee, and bowl. Everything else was generated with Marble. Not sure how to do this yourself? NVIDIA wrote a technical blog with full instructions. You will learn how to: 1. Export scenes from World Labs' Marble as Gaussian splats 2. Convert to USD using NVIDIA Omniverse NuRec 3. Import into NVIDIA Isaac Sim 4. Add a robot and run the simulation Read the guide ➡️ #WorldModels #Robotics
Jonathan Stephens13,783 Aufrufe • vor 10 Monaten
