正在加载视频...

视频加载失败

返回首页

Teaching our open humanoid ROBOTIS AI Sapiens to dance! 🤖💃 We extracted 3D motion from raw 2D video (no MoCap device!) and trained it in NVIDIA Isaac Sim. Thanks to high-precision DYNAMIXEL-Q, we achieved an ultra-low Sim2Real gap for fluid movement! - Video:

passionvirus's profile picture

Yoonseok Pyo

1,244 subscribers

158,164 次观看 • 3 个月前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

After two years in stealth mode, we're thrilled to unveil Mentee Robotics and our humanoid robot, Mentee Robotics! With AI integration at every layer, from Sim2Real machine learning to NeRF-based algorithms and LLMs, we've achieved a complete end-to-end cycle for tasks. This is just the beginning and we can’t wait to share more
2:26

After two years in stealth mode, we're thrilled to unveil Mentee Robotics and our humanoid robot, Mentee Robotics! With AI integration at every layer, from Sim2Real machine learning to NeRF-based algorithms and LLMs, we've achieved a complete end-to-end cycle for tasks. This is just the beginning and we can’t wait to share more

Amnon Shashua

223,942 次观看 • 2 年前

Mobile-TeleVision follows our previous idea on upper-body and lower-body separation for control: 1⃣We train RL for lower-body locomotion and standing; 2⃣We use IK and retargeting directly for upper-body for high-precision and smooth manipulation. We avoid using RL for upper-body motion tracking. This is because fundamentally it is hard to avoid errors with RL tracking, and sim2real gap brings another gap. At this point, we still find using IK is more practical and precise.
1:15

Mobile-TeleVision follows our previous idea on upper-body and lower-body separation for control: 1⃣We train RL for lower-body locomotion and standing; 2⃣We use IK and retargeting directly for upper-body for high-precision and smooth manipulation. We avoid using RL for upper-body motion tracking. This is because fundamentally it is hard to avoid errors with RL tracking, and sim2real gap brings another gap. At this point, we still find using IK is more practical and precise.

Xiaolong Wang

32,662 次观看 • 1 年前

Physics AI 🌊 Jensen just unveiled the NVIDIA Isaac GR00T Reference Humanoid Robot at GTC Taipei--the world’s first open-source humanoid robot reference design Built on Jetson Thor edge computing and the Isaac GR00T open platform, it features a Unitree H2 Plus body with Sharpa five-finger dexterous hands and 75 degrees of freedom. Even before Physical AI truly explodes, NVIDIA has built a complete end-to-end full-stack infrastructure: >chips (Jetson Thor + Blackwell), >synthetic data (GR00T-Dreams), >world models (Cosmos), >foundation models (GR00T N-series VLA), >high-fidelity simulation platforms (Isaac Sim + Isaac Lab + Omniverse). Dozens of humanoid robot companies ,including Agility Robotics, Aglie robotics,Boston Dynamics, Figure, 1X, NEURA Robotics, XPENG Robotics, and more are already running large-scale daily simulation training on this platform, rapidly iterating perception, reasoning, and actions in virtual environments, speeding up the jump from lab prototypes to real factory deployment. By providing an open reference design and full-stack tools, NVIDIA has dramatically lowered the barrier, letting every robotics team join the Physical AI wave with low cost and high efficiency.
4:31

Physics AI 🌊 Jensen just unveiled the NVIDIA Isaac GR00T Reference Humanoid Robot at GTC Taipei--the world’s first open-source humanoid robot reference design Built on Jetson Thor edge computing and the Isaac GR00T open platform, it features a Unitree H2 Plus body with Sharpa five-finger dexterous hands and 75 degrees of freedom. Even before Physical AI truly explodes, NVIDIA has built a complete end-to-end full-stack infrastructure: >chips (Jetson Thor + Blackwell), >synthetic data (GR00T-Dreams), >world models (Cosmos), >foundation models (GR00T N-series VLA), >high-fidelity simulation platforms (Isaac Sim + Isaac Lab + Omniverse). Dozens of humanoid robot companies ,including Agility Robotics, Aglie robotics,Boston Dynamics, Figure, 1X, NEURA Robotics, XPENG Robotics, and more are already running large-scale daily simulation training on this platform, rapidly iterating perception, reasoning, and actions in virtual environments, speeding up the jump from lab prototypes to real factory deployment. By providing an open reference design and full-stack tools, NVIDIA has dramatically lowered the barrier, letting every robotics team join the Physical AI wave with low cost and high efficiency.

CyberRobo

17,092 次观看 • 3 个月前

In our latest video, we completely disassemble the Unitree G1 humanoid robot, taking an in depth look at its engineering and design. We analyze the robot's actuators, motors, and thermal management systems to understand how Unitree balances performance, cost, and manufacturability. A special thanks to Humanoid Scott for joining the conversation and sharing his insights.
4:38

In our latest video, we completely disassemble the Unitree G1 humanoid robot, taking an in depth look at its engineering and design. We analyze the robot's actuators, motors, and thermal management systems to understand how Unitree balances performance, cost, and manufacturability. A special thanks to Humanoid Scott for joining the conversation and sharing his insights.

Munro Live

61,869 次观看 • 2 个月前

This is the last year I create content in 2D. To prove it I’m releasing an 11-minute 4DGS video in partnership with 4DV.ai where every single shot is dynamic 3D. Over the next 12–18 months, I'll be moving my filming exclusively to dynamic 3D. Thank you so much to NVIDIA AI and Dell Technologies for making this video possible.
0:34

This is the last year I create content in 2D. To prove it I’m releasing an 11-minute 4DGS video in partnership with 4DV.ai where every single shot is dynamic 3D. Over the next 12–18 months, I'll be moving my filming exclusively to dynamic 3D. Thank you so much to NVIDIA AI and Dell Technologies for making this video possible.

Radiance Fields

23,993 次观看 • 1 个月前

Take a peek behind the scenes of our Snoop Dogg music video to see how we bring our shots to life in our projects! For the chariot scene, we used AI to generate movement between key frames and seamlessly connect them to create smooth animations. Thanks to Yves Dalbiez for creating this shot and breakdown!
2:19

Take a peek behind the scenes of our Snoop Dogg music video to see how we bring our shots to life in our projects! For the chariot scene, we used AI to generate movement between key frames and seamlessly connect them to create smooth animations. Thanks to Yves Dalbiez for creating this shot and breakdown!

The Dor Brothers

64,751 次观看 • 1 年前

Long video generation is a systems problem. Introducing LongLive-2.0 from NVIDIA Research: an end-to-end NVFP4 training and inference system for long video generation. Low-precision deployment often relies on post-training quantization, creating a gap between how models are trained and how they run. LongLive-2.0 aligns NVFP4-aware training, distillation, and W4A4 inference, maintaining strong benchmark quality while improving speed and memory efficiency.
0:59

Long video generation is a systems problem. Introducing LongLive-2.0 from NVIDIA Research: an end-to-end NVFP4 training and inference system for long video generation. Low-precision deployment often relies on post-training quantization, creating a gap between how models are trained and how they run. LongLive-2.0 aligns NVFP4-aware training, distillation, and W4A4 inference, maintaining strong benchmark quality while improving speed and memory efficiency.

NVIDIA AI

60,690 次观看 • 3 个月前

Update: We hired Tallis to create this video for our new high end office product in San Ramon, CA - and he crushed it! We send the video to prospective tenants, and we are seeing an even greater increase in tours! Love to pay it forward - he can do the same for your building, anywhere in America !!!
1:31

Update: We hired Tallis to create this video for our new high end office product in San Ramon, CA - and he crushed it! We send the video to prospective tenants, and we are seeing an even greater increase in tours! Love to pay it forward - he can do the same for your building, anywhere in America !!!

StripMallGuy

35,213 次观看 • 3 个月前

Any movement, any subject, any environment. 📸 That's the creative freedom of Meshcapade's Markerless MoCapade. 🚀 We deliver the industry's most accurate 3D motion and body shape data from any single video source. Our AI transforms footage—from subtle social gestures to intense dynamic movements—into precise, high-fidelity 3D data ready for immediate integration (FBX, GLB). Cut the studio time. Get the motion you need, right now. It doesn't get SMPL-er than this. 😉 #MarkerlessMocap #GameDev #3DAnimation #Meshcapade #SMPLX
0:36

Any movement, any subject, any environment. 📸 That's the creative freedom of Meshcapade's Markerless MoCapade. 🚀 We deliver the industry's most accurate 3D motion and body shape data from any single video source. Our AI transforms footage—from subtle social gestures to intense dynamic movements—into precise, high-fidelity 3D data ready for immediate integration (FBX, GLB). Cut the studio time. Get the motion you need, right now. It doesn't get SMPL-er than this. 😉 #MarkerlessMocap #GameDev #3DAnimation #Meshcapade #SMPLX

Meshcapade

31,646 次观看 • 10 个月前

Historically, RL policies for robots have been trained in synthetic, untextured environments, limiting perceptive policies to depth images where the sim-to-real gap is manageable. RGB has always had more potential, but leveraging it to train policies in simulation remained an open problem. Partnering with Niantic Spatial 🌎 and NVIDIA Robotics, we built a pipeline that addresses exactly that. We can now scan a real deployment site with off-the-shelf hardware, reconstruct it into a photorealistic Gaussian splat, and run massively parallel RL training. The policies trained in our Gym environment then transfer zero-shot to the real robot and environments they were trained for. This enables faster deployment of more capable and robust policies for the end user. The new resulting capabilities are a big step towards solving sim-to-real and also apply well beyond navigation. Read the full technical breakdown on our blog; link in the comments. #HumanoidRobots #Flexion #NianticSpatial #NVIDIA
1:21

Historically, RL policies for robots have been trained in synthetic, untextured environments, limiting perceptive policies to depth images where the sim-to-real gap is manageable. RGB has always had more potential, but leveraging it to train policies in simulation remained an open problem. Partnering with Niantic Spatial 🌎 and NVIDIA Robotics, we built a pipeline that addresses exactly that. We can now scan a real deployment site with off-the-shelf hardware, reconstruct it into a photorealistic Gaussian splat, and run massively parallel RL training. The policies trained in our Gym environment then transfer zero-shot to the real robot and environments they were trained for. This enables faster deployment of more capable and robust policies for the end user. The new resulting capabilities are a big step towards solving sim-to-real and also apply well beyond navigation. Read the full technical breakdown on our blog; link in the comments. #HumanoidRobots #Flexion #NianticSpatial #NVIDIA

Flexion

72,692 次观看 • 2 个月前

my V2 system is by far the most high quality and most efficient AI video system right now we have perfected: > ultra realistic voice (no robotic AI voice) > ultra realistic visuals > ultra realistic movement and no it doesn't take 1 hour of editing to get a decent result. this video took me 7 minutes to make and costs 50 cents the system is built for scale and quality. there's little to no manual work you're actually doing here. we're using these videos across every industry and seeing insane results. ecom, saas, info, everything...
0:38

my V2 system is by far the most high quality and most efficient AI video system right now we have perfected: > ultra realistic voice (no robotic AI voice) > ultra realistic visuals > ultra realistic movement and no it doesn't take 1 hour of editing to get a decent result. this video took me 7 minutes to make and costs 50 cents the system is built for scale and quality. there's little to no manual work you're actually doing here. we're using these videos across every industry and seeing insane results. ecom, saas, info, everything...

Miko

16,850 次观看 • 5 个月前

We are pleased to announce our strategic collaboration with Holoworld Holoworld AI / AVA $AVA , we are working closely with one of the best AI agent teams in the industry @HolowordAI . An exclusive integration will be made with Holoworld for ActSwap which will allow access to prompts from Holoworld agents like AVA. $ACT ecosystem agents will soon be brought to life through 2D/3D avatars, voice and media generation thanks to Holoworld framework : the audiovisual layer for AI agents.
0:26

We are pleased to announce our strategic collaboration with Holoworld Holoworld AI / AVA $AVA , we are working closely with one of the best AI agent teams in the industry @HolowordAI . An exclusive integration will be made with Holoworld for ActSwap which will allow access to prompts from Holoworld agents like AVA. $ACT ecosystem agents will soon be brought to life through 2D/3D avatars, voice and media generation thanks to Holoworld framework : the audiovisual layer for AI agents.

Act I : The AI Prophecy

185,163 次观看 • 1 年前

Humanoids need a single, generalist control policy for all of their physical tasks, not a new one for every new chore or demo. A policy for walking can't dance. A policy for dancing can't support mowing the lawn. We need to scale up humanoid control for diverse behaviors, just like we did with language and video models. Watch this humanoid, supported by a single control policy, to start its day, mow the lawn, jog, dance, and even hold the elevator for a colleague. This is the "System 1" for humanoids—a robust, natural humanoid controller that will support general-purpose autonomy. Project Page: #Robotics #AI #NVIDIA #SONIC #FoundationModel 🤖🔥
1:33

Humanoids need a single, generalist control policy for all of their physical tasks, not a new one for every new chore or demo. A policy for walking can't dance. A policy for dancing can't support mowing the lawn. We need to scale up humanoid control for diverse behaviors, just like we did with language and video models. Watch this humanoid, supported by a single control policy, to start its day, mow the lawn, jog, dance, and even hold the elevator for a colleague. This is the "System 1" for humanoids—a robust, natural humanoid controller that will support general-purpose autonomy. Project Page: #Robotics #AI #NVIDIA #SONIC #FoundationModel 🤖🔥

Zhengyi “Zen” Luo

15,982 次观看 • 10 个月前

Our website now supports Kling Motion Control. Today, creating an AI influencer video has become incredibly simple. You can first use Nano Banana Pro to generate realistic images, then use Kling Motion Control to replace the motion and quickly create viral-style videos for social media. We also provide a selection of popular online dance motions with controllable presets for you to use. If you're interested, feel free to try it on our website:
0:23

Our website now supports Kling Motion Control. Today, creating an AI influencer video has become incredibly simple. You can first use Nano Banana Pro to generate realistic images, then use Kling Motion Control to replace the motion and quickly create viral-style videos for social media. We also provide a selection of popular online dance motions with controllable presets for you to use. If you're interested, feel free to try it on our website:

underwood

15,467 次观看 • 6 个月前

🤖 Let’s MEET the team! 🤝 We are excited to introduce our full #AI team who will be leading the project from all aspects. Take a moment to watch the video to get to know our team and learn more about what we have in store for #AIPad 🕸 #Crypto #IDO
2:20

🤖 Let’s MEET the team! 🤝 We are excited to introduce our full #AI team who will be leading the project from all aspects. Take a moment to watch the video to get to know our team and learn more about what we have in store for #AIPad 🕸 #Crypto #IDO

AIPAD.TECH

66,565 次观看 • 3 年前

EDIT MOTION IN VIDEOS!!! Quit prompting and start directing I've been shouting for YEARS about 3D as the control layer. Here it is, signs of life of our Universal Video Editor!!! The workflow is: take your video, capture with comic 4, edit with the motion editor, re-render with your favorite video to video model (e.g., Runway and Google Gemini have good ones) Here we have a fashion clip where we wanted our actress to high step and show a bit more pizzaz - but shoot day is over, and traditionally it would cost many thousands to reshoot. Watch the video to see how it works instead in cartwheel. cc Logan Kilpatrick Cristóbal Valenzuela 👀
0:20

EDIT MOTION IN VIDEOS!!! Quit prompting and start directing I've been shouting for YEARS about 3D as the control layer. Here it is, signs of life of our Universal Video Editor!!! The workflow is: take your video, capture with comic 4, edit with the motion editor, re-render with your favorite video to video model (e.g., Runway and Google Gemini have good ones) Here we have a fashion clip where we wanted our actress to high step and show a bit more pizzaz - but shoot day is over, and traditionally it would cost many thousands to reshoot. Watch the video to see how it works instead in cartwheel. cc Logan Kilpatrick Cristóbal Valenzuela 👀

Andrew Carr 🤸

11,622 次观看 • 3 个月前

Generative Rendering: Controllable 4D-Guided Video Generation with 2D Diffusion Models paper page: Traditional 3D content creation tools empower users to bring their imagination to life by giving them direct control over a scene's geometry, appearance, motion, and camera path. Creating computer-generated videos, however, is a tedious manual process, which can be automated by emerging text-to-video diffusion models. Despite great promise, video diffusion models are difficult to control, hindering a user to apply their own creativity rather than amplifying it. To address this challenge, we present a novel approach that combines the controllability of dynamic 3D meshes with the expressivity and editability of emerging diffusion models. For this purpose, our approach takes an animated, low-fidelity rendered mesh as input and injects the ground truth correspondence information obtained from the dynamic mesh into various stages of a pre-trained text-to-image generation model to output high-quality and temporally consistent frames. We demonstrate our approach on various examples where motion can be obtained by animating rigged assets or changing the camera path.
0:23

Generative Rendering: Controllable 4D-Guided Video Generation with 2D Diffusion Models paper page: Traditional 3D content creation tools empower users to bring their imagination to life by giving them direct control over a scene's geometry, appearance, motion, and camera path. Creating computer-generated videos, however, is a tedious manual process, which can be automated by emerging text-to-video diffusion models. Despite great promise, video diffusion models are difficult to control, hindering a user to apply their own creativity rather than amplifying it. To address this challenge, we present a novel approach that combines the controllability of dynamic 3D meshes with the expressivity and editability of emerging diffusion models. For this purpose, our approach takes an animated, low-fidelity rendered mesh as input and injects the ground truth correspondence information obtained from the dynamic mesh into various stages of a pre-trained text-to-image generation model to output high-quality and temporally consistent frames. We demonstrate our approach on various examples where motion can be obtained by animating rigged assets or changing the camera path.

AK

117,068 次观看 • 2 年前

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 →
0:47

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

119,356 次观看 • 1 个月前