正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

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:show more

Ilir Aliu

53,912 subscribers

12,950 次观看 • 1 个月前 •via X (Twitter)

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

Robotics keeps hitting the same wall. Single task RL works, but... it does not scale to hundreds of tasks or new embodiments. This new paper looks like a real step toward fixing that. The team introduces MMBench, a benchmark with 200 tasks across many domains and robots, and Newt, a language conditioned world model trained online across all 200 tasks at once. The simple idea behind Newt: The model learns from demos to get the right priors It trains across many tasks through online interaction It uses language to ground the goal It adapts fast when a new task shows up What stood out to me: ✅ One model trained on 200 tasks at the same time ✅ Language conditioned control for both states and RGB ✅ Better data efficiency than strong baselines ✅ Strong open loop control ✅ Fast adaptation to new tasks and embodiments ✅ Full release of 200 checkpoints, 4000 demos, code, and benchmark This is a good push toward general control instead of one model per task. If you want the full paper: Project page: —- Weekly robotics and AI insights. Subscribe free:

Robotics keeps hitting the same wall. Single task RL works, but... it does not scale to hundreds of tasks or new embodiments. This new paper looks like a real step toward fixing that. The team introduces MMBench, a benchmark with 200 tasks across many domains and robots, and Newt, a language conditioned world model trained online across all 200 tasks at once. The simple idea behind Newt: The model learns from demos to get the right priors It trains across many tasks through online interaction It uses language to ground the goal It adapts fast when a new task shows up What stood out to me: ✅ One model trained on 200 tasks at the same time ✅ Language conditioned control for both states and RGB ✅ Better data efficiency than strong baselines ✅ Strong open loop control ✅ Fast adaptation to new tasks and embodiments ✅ Full release of 200 checkpoints, 4000 demos, code, and benchmark This is a good push toward general control instead of one model per task. If you want the full paper: Project page: —- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

70,090 次观看 • 9 个月前

A Gaussian Splat can become a world where Robots and AI agents can act. In our latest OVER Research experiment, we placed a robot inside a real-world 3D capture, with a VLM making decisions based on what it sees. At every step, the robot holds a pose in the reconstruction, gets a newly rendered view of the environment, takes an action, moves, and sees the world again from its new position. Why does this matter? Because 3D captures can become more than reconstructions to explore. They can become environments where embodied AI and robots can navigate, act, be evaluated and eventually train across real-world spaces at scale. Capture a place once. Then turn it into a world where AI and Robots can act. The full experiment, including what we discovered once we actually put the loop to the test:

A Gaussian Splat can become a world where Robots and AI agents can act. In our latest OVER Research experiment, we placed a robot inside a real-world 3D capture, with a VLM making decisions based on what it sees. At every step, the robot holds a pose in the reconstruction, gets a newly rendered view of the environment, takes an action, moves, and sees the world again from its new position. Why does this matter? Because 3D captures can become more than reconstructions to explore. They can become environments where embodied AI and robots can navigate, act, be evaluated and eventually train across real-world spaces at scale. Capture a place once. Then turn it into a world where AI and Robots can act. The full experiment, including what we discovered once we actually put the loop to the test:

Over the Reality 🌐

14,946 次观看 • 10 天前

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

Jonathan Stephens

17,712 次观看 • 8 个月前

Super excited about Hydra-0 from Hongyu Li and team! The key idea is to use flow as a shared visual interface across embodiments/objects for controllable video generation, allowing a single generalist world model to learn from human, handheld-gripper, and robot interaction data. My favorite result is the video below: start from a real video of a human doing the task (left), extract the desired object flow, and condition the model on that flow (right). The model then hallucinates a plausible robot motion that could produce the same object motion. Very cool glimpse of how a generalist world model can bridge human demonstrations and robot control.

Super excited about Hydra-0 from Hongyu Li and team! The key idea is to use flow as a shared visual interface across embodiments/objects for controllable video generation, allowing a single generalist world model to learn from human, handheld-gripper, and robot interaction data. My favorite result is the video below: start from a real video of a human doing the task (left), extract the desired object flow, and condition the model on that flow (right). The model then hallucinates a plausible robot motion that could produce the same object motion. Very cool glimpse of how a generalist world model can bridge human demonstrations and robot control.

Yunzhu Li

10,509 次观看 • 21 天前

Robora Sim: A PyBullet-Powered Environment for Learning Robotic Physical Intelligence We are currently building our Robora simulation environment setup for our sim based learning, leveraging PyBullet, an industry-standard physics engine widely used in AI-driven robotics research and development. The environment is optimized with GPU-accelerated learning algorithms, enabling high-speed imitation learning and reinforcement learning within a safe and controlled virtual setup before shipping out to real world. This simulation platform allows our models to learn, adapt, and generalize across different robot morphologies, terrain types and task objectives - all before deployment to the real world. At it's core, the system combines a VLA-powered high-level planner with low-level motion control algorithms, working cohesively to produce emergent, physically intelligent behaviors. This synergy between simulation, learning, and real-world transfer marks a major step forward in our pursuit of adaptive and intelligent robotic systems. Through advanced domain randomization and synthetic data generation, the Robora Simulation Environment ensures that policies trained in simulation transfer effectively to real-world robots, minimizing the sim-to-real gap. Moreover, users will be able to test and integrate their own hardware kits within selected simulation environments in the Robora Dapp, ensuring seamless compatibility and safer real-world implementation.

Robora Sim: A PyBullet-Powered Environment for Learning Robotic Physical Intelligence We are currently building our Robora simulation environment setup for our sim based learning, leveraging PyBullet, an industry-standard physics engine widely used in AI-driven robotics research and development. The environment is optimized with GPU-accelerated learning algorithms, enabling high-speed imitation learning and reinforcement learning within a safe and controlled virtual setup before shipping out to real world. This simulation platform allows our models to learn, adapt, and generalize across different robot morphologies, terrain types and task objectives - all before deployment to the real world. At it's core, the system combines a VLA-powered high-level planner with low-level motion control algorithms, working cohesively to produce emergent, physically intelligent behaviors. This synergy between simulation, learning, and real-world transfer marks a major step forward in our pursuit of adaptive and intelligent robotic systems. Through advanced domain randomization and synthetic data generation, the Robora Simulation Environment ensures that policies trained in simulation transfer effectively to real-world robots, minimizing the sim-to-real gap. Moreover, users will be able to test and integrate their own hardware kits within selected simulation environments in the Robora Dapp, ensuring seamless compatibility and safer real-world implementation.

Robora

23,489 次观看 • 11 个月前

TESLA HALTED MODEL S AND MODEL X PRODUCTION TO BUILD AN ARMY OF OPTIMUS ROBOTS The Fremont assembly line was torn down in 46 days. In its place, Tesla is building a line for humanoid production, aiming for a million units a year A humanoid robot is a body shaped like a human. Physical AI is the intelligence that controls that body Walking and making coffee is often just imitation learning from a scripted routine. But once the environment shifts, the learned trick stops working Language models had the entire internet to train on. Robotics has nothing close to that scale of data, which is why one giant brain hasn't worked for anyone yet The industry is moving toward modularity instead - separate models for vision, movement, and planning, each improved on its own The real question is no longer whether a robot can move impressively. It's whether it can pull its sensors into one picture of the world and adapt to whatever wasn't scripted for it

TESLA HALTED MODEL S AND MODEL X PRODUCTION TO BUILD AN ARMY OF OPTIMUS ROBOTS The Fremont assembly line was torn down in 46 days. In its place, Tesla is building a line for humanoid production, aiming for a million units a year A humanoid robot is a body shaped like a human. Physical AI is the intelligence that controls that body Walking and making coffee is often just imitation learning from a scripted routine. But once the environment shifts, the learned trick stops working Language models had the entire internet to train on. Robotics has nothing close to that scale of data, which is why one giant brain hasn't worked for anyone yet The industry is moving toward modularity instead - separate models for vision, movement, and planning, each improved on its own The real question is no longer whether a robot can move impressively. It's whether it can pull its sensors into one picture of the world and adapt to whatever wasn't scripted for it

iamigorekk

22,205 次观看 • 25 天前

Something big is happening in robotics - and it’s hiding in plain sight. This post is not about dancing robots but in the data that powers them. Open robotics datasets have exploded this year, turning the field into a more scalable and collaborative ecosystem. In just two years, Hugging Face datasets grew from 11k to over 600k - and robotics is by far the fastest-growing segment. We went from 1k robotics datasets in 2024 to 27k in 2025! For comparison, text generation, the second-largest category, has only around 5k datasets in 2025. That gap is massive. Open datasets are important because robotics lives and dies by real-world robot data - video, actions, sensors, failures. By making this data easy to upload, reuse, and benchmark, researchers, startups, and large players are now releasing real-robot datasets that would have stayed locked inside labs just a few years ago. Major contributors include NVIDIA, LeRobot initiative, and a rapidly growing maker community. This surge is also enabled by cheaper video storage, better tooling, and an open-source AI culture now spilling into the physical world. And it really matters: open robotics data dramatically lowers entry barriers, accelerates learning-by-doing, and speeds up progress toward generalist and humanoid robots. Robotics won’t scale through hardware alone - but to a large extent through shared data. Viz below from AI World - link to the story and more viz/filters in comment.

Something big is happening in robotics - and it’s hiding in plain sight. This post is not about dancing robots but in the data that powers them. Open robotics datasets have exploded this year, turning the field into a more scalable and collaborative ecosystem. In just two years, Hugging Face datasets grew from 11k to over 600k - and robotics is by far the fastest-growing segment. We went from 1k robotics datasets in 2024 to 27k in 2025! For comparison, text generation, the second-largest category, has only around 5k datasets in 2025. That gap is massive. Open datasets are important because robotics lives and dies by real-world robot data - video, actions, sensors, failures. By making this data easy to upload, reuse, and benchmark, researchers, startups, and large players are now releasing real-robot datasets that would have stayed locked inside labs just a few years ago. Major contributors include NVIDIA, LeRobot initiative, and a rapidly growing maker community. This surge is also enabled by cheaper video storage, better tooling, and an open-source AI culture now spilling into the physical world. And it really matters: open robotics data dramatically lowers entry barriers, accelerates learning-by-doing, and speeds up progress toward generalist and humanoid robots. Robotics won’t scale through hardware alone - but to a large extent through shared data. Viz below from AI World - link to the story and more viz/filters in comment.

Pierre-Alexandre Balland

186,094 次观看 • 8 个月前

If you think OpenAI Sora is a creative toy like DALLE, ... think again. Sora is a data-driven physics engine. It is a simulation of many worlds, real or fantastical. The simulator learns intricate rendering, "intuitive" physics, long-horizon reasoning, and semantic grounding, all by some denoising and gradient maths. I won't be surprised if Sora is trained on lots of synthetic data using Unreal Engine 5. It has to be! Let's breakdown the following video. Prompt: "Photorealistic closeup video of two pirate ships battling each other as they sail inside a cup of coffee." - The simulator instantiates two exquisite 3D assets: pirate ships with different decorations. Sora has to solve text-to-3D implicitly in its latent space. - The 3D objects are consistently animated as they sail and avoid each other's paths. - Fluid dynamics of the coffee, even the foams that form around the ships. Fluid simulation is an entire sub-field of computer graphics, which traditionally requires very complex algorithms and equations. - Photorealism, almost like rendering with raytracing. - The simulator takes into account the small size of the cup compared to oceans, and applies tilt-shift photography to give a "minuscule" vibe. - The semantics of the scene does not exist in the real world, but the engine still implements the correct physical rules that we expect. Next up: add more modalities and conditioning, then we have a full data-driven UE that will replace all the hand-engineered graphics pipelines.

If you think OpenAI Sora is a creative toy like DALLE, ... think again. Sora is a data-driven physics engine. It is a simulation of many worlds, real or fantastical. The simulator learns intricate rendering, "intuitive" physics, long-horizon reasoning, and semantic grounding, all by some denoising and gradient maths. I won't be surprised if Sora is trained on lots of synthetic data using Unreal Engine 5. It has to be! Let's breakdown the following video. Prompt: "Photorealistic closeup video of two pirate ships battling each other as they sail inside a cup of coffee." - The simulator instantiates two exquisite 3D assets: pirate ships with different decorations. Sora has to solve text-to-3D implicitly in its latent space. - The 3D objects are consistently animated as they sail and avoid each other's paths. - Fluid dynamics of the coffee, even the foams that form around the ships. Fluid simulation is an entire sub-field of computer graphics, which traditionally requires very complex algorithms and equations. - Photorealism, almost like rendering with raytracing. - The simulator takes into account the small size of the cup compared to oceans, and applies tilt-shift photography to give a "minuscule" vibe. - The semantics of the scene does not exist in the real world, but the engine still implements the correct physical rules that we expect. Next up: add more modalities and conditioning, then we have a full data-driven UE that will replace all the hand-engineered graphics pipelines.

Jim Fan

6,183,365 次观看 • 2 年前

Multi-robot learning is getting a serious boost! 📚 Researchers have extended Isaac Lab to train heterogeneous multi-agent robotic policies at scale. The new framework supports high-resolution physics, GPU-accelerated simulation, and both homogeneous and heterogeneous agents working together on coordination tasks. They benchmarked different approaches (MAPPO: Multi-Agent Proximal Policy Optimization and HAPPO: Heterogeneous Agent PPO) across six challenging scenarios and showed that large-scale multi-robot training is not only feasible, but efficient. It’s an important step for real-world robotic collaboration, where teams of robots need to coordinate, split tasks, adapt roles, and interact dynamically, not just operate as identical clones. The code is open-source, and it pushes Isaac Lab closer to what robotics actually needs: scalable, physics-driven environments where many different robots can learn to work together. Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Multi-robot learning is getting a serious boost! 📚 Researchers have extended Isaac Lab to train heterogeneous multi-agent robotic policies at scale. The new framework supports high-resolution physics, GPU-accelerated simulation, and both homogeneous and heterogeneous agents working together on coordination tasks. They benchmarked different approaches (MAPPO: Multi-Agent Proximal Policy Optimization and HAPPO: Heterogeneous Agent PPO) across six challenging scenarios and showed that large-scale multi-robot training is not only feasible, but efficient. It’s an important step for real-world robotic collaboration, where teams of robots need to coordinate, split tasks, adapt roles, and interact dynamically, not just operate as identical clones. The code is open-source, and it pushes Isaac Lab closer to what robotics actually needs: scalable, physics-driven environments where many different robots can learn to work together. Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

38,997 次观看 • 9 个月前

Check out this Stereo4D paper from Google DeepMind. It's a pretty clever approach to a persistent problem in computer vision -- getting good training data for how things move in 3D. The key insight is using VR180 videos -- those stereo fisheye videos we launched back in 2017 for YouTubeVR. It was always clear that structured stereo datasets would be valuable for computer vision -- and we launched some powerful VR tools with it back in 2017 (link below). But what's the game changer now in 2024 is the scale -- they're providing 110K high quality clips :-) That's the kind of massive, real-world AI dataset that was just a dream back then! They're using it to train this model called DynaDUSt3R that can predict both 3D structure and motion from video frames. Which means it tracks how objects move between frames while simultaneously reconstructing their 3D shape. And given we're dealing with real stereoscopic content, results are notably better than synthetic data, giving you a faithful rendition of the real-world with a diverse set of subject matter. It's one of those through lines when tackling a timeless mission like mapping the world or spatial computing -- VR content created for immersion becoming the foundation for teaching machines to understand how the world moves. Sometimes innovation chains together in unexpected ways! Links to projects below⛓️

Check out this Stereo4D paper from Google DeepMind. It's a pretty clever approach to a persistent problem in computer vision -- getting good training data for how things move in 3D. The key insight is using VR180 videos -- those stereo fisheye videos we launched back in 2017 for YouTubeVR. It was always clear that structured stereo datasets would be valuable for computer vision -- and we launched some powerful VR tools with it back in 2017 (link below). But what's the game changer now in 2024 is the scale -- they're providing 110K high quality clips :-) That's the kind of massive, real-world AI dataset that was just a dream back then! They're using it to train this model called DynaDUSt3R that can predict both 3D structure and motion from video frames. Which means it tracks how objects move between frames while simultaneously reconstructing their 3D shape. And given we're dealing with real stereoscopic content, results are notably better than synthetic data, giving you a faithful rendition of the real-world with a diverse set of subject matter. It's one of those through lines when tackling a timeless mission like mapping the world or spatial computing -- VR content created for immersion becoming the foundation for teaching machines to understand how the world moves. Sometimes innovation chains together in unexpected ways! Links to projects below⛓️

Bilawal Sidhu

68,515 次观看 • 1 年前

Your model crushed the benchmark. Then it couldn’t pick up a cup. That’s the reality nobody talks about. You train in simulation, it falls apart on real hardware. You collect real-world data instead (months of teleop, physical setups, safety protocols) and still can’t scale it. Meanwhile, your model outgrows every available benchmark, and you have no way to know if you’re actually getting better. That costs you iteration speed. Engineering hours. And confidence in every decision about when to ship. This year I’m collaborating with Lightwheel to cover robotics and embodied AI at NVIDIA GTC — Booth 1406, March 16–19, San Jose. Live demos of what happens when the physics ACTUALLY match. Worth a stop. I’ll be there. Saying hello to Steve Xie 👋

Your model crushed the benchmark. Then it couldn’t pick up a cup. That’s the reality nobody talks about. You train in simulation, it falls apart on real hardware. You collect real-world data instead (months of teleop, physical setups, safety protocols) and still can’t scale it. Meanwhile, your model outgrows every available benchmark, and you have no way to know if you’re actually getting better. That costs you iteration speed. Engineering hours. And confidence in every decision about when to ship. This year I’m collaborating with Lightwheel to cover robotics and embodied AI at NVIDIA GTC — Booth 1406, March 16–19, San Jose. Live demos of what happens when the physics ACTUALLY match. Worth a stop. I’ll be there. Saying hello to Steve Xie 👋

Ilir Aliu

10,575 次观看 • 6 个月前

SOMEONE FROM TOKYO IS MAPPING BIRD LANGUAGE INTO REAL DATA PATTERNS AND THE VISUALIZATION LOOKS LIKE A NEURAL NETWORK DREAMING Every bird sound - frequency, duration, amplitude and modulation - gets converted into 3D space coordinates and rendered as a cluster of colored points in real time through Deepen AI. A microphone captures the acoustic signal, FFT breaks it down into frequency components from 1 kHz to 8 kHz where most birds communicate, an ML model classifies the pattern and the mapped data hits a graph where each call type gets its own color and position in space. Birds have a vocal repertoire of 5 to 200+ unique signals depending on the species - and every signal carries different information about predators, food, territory and mates that humans simply can't hear. The same technology that decodes bird language detects anomalies in industrial machinery, cardiac rhythms and structural vibrations in buildings - he just trained it on nature first.

SOMEONE FROM TOKYO IS MAPPING BIRD LANGUAGE INTO REAL DATA PATTERNS AND THE VISUALIZATION LOOKS LIKE A NEURAL NETWORK DREAMING Every bird sound - frequency, duration, amplitude and modulation - gets converted into 3D space coordinates and rendered as a cluster of colored points in real time through Deepen AI. A microphone captures the acoustic signal, FFT breaks it down into frequency components from 1 kHz to 8 kHz where most birds communicate, an ML model classifies the pattern and the mapped data hits a graph where each call type gets its own color and position in space. Birds have a vocal repertoire of 5 to 200+ unique signals depending on the species - and every signal carries different information about predators, food, territory and mates that humans simply can't hear. The same technology that decodes bird language detects anomalies in industrial machinery, cardiac rhythms and structural vibrations in buildings - he just trained it on nature first.

Cortex

11,128 次观看 • 3 个月前

This work makes a humanoid robot do simple parkour moves by looking with a depth camera and choosing the right move on the fly. The big deal is that it turns lots of small human moves into long, real-time robot behavior, without hand-coding every transition or retraining for each new course. A humanoid robot is usually good at steady walking, but it often fails when it has to do fast moves like jumping up, vaulting, or rolling, and then keep going to the next obstacle. The hard part is that you cannot easily collect training data for every possible obstacle shape, distance, and mistake, so robots end up learning a few moves that only work in a narrow setup. This work starts from short clips of real human parkour moves, like stepping over, vaulting, climbing, and rolling. It uses motion matching, which is basically a smart “pick the next clip that fits best right now” search, to stitch those short clips into a long, smooth plan that looks like a human doing a whole course. Then it trains a controller with reinforcement learning (RL), which means the robot learns by trial and error to copy that plan while staying balanced and not falling. After training separate expert controllers for different moves, it compresses them into 1 controller that uses only onboard depth sensing and a simple “go this fast in this direction” command. In real tests on a Unitree G1 humanoid, it can clear multiple obstacles in a row, adapt when obstacles get moved, and climb a wall up to 1.25m.

This work makes a humanoid robot do simple parkour moves by looking with a depth camera and choosing the right move on the fly. The big deal is that it turns lots of small human moves into long, real-time robot behavior, without hand-coding every transition or retraining for each new course. A humanoid robot is usually good at steady walking, but it often fails when it has to do fast moves like jumping up, vaulting, or rolling, and then keep going to the next obstacle. The hard part is that you cannot easily collect training data for every possible obstacle shape, distance, and mistake, so robots end up learning a few moves that only work in a narrow setup. This work starts from short clips of real human parkour moves, like stepping over, vaulting, climbing, and rolling. It uses motion matching, which is basically a smart “pick the next clip that fits best right now” search, to stitch those short clips into a long, smooth plan that looks like a human doing a whole course. Then it trains a controller with reinforcement learning (RL), which means the robot learns by trial and error to copy that plan while staying balanced and not falling. After training separate expert controllers for different moves, it compresses them into 1 controller that uses only onboard depth sensing and a simple “go this fast in this direction” command. In real tests on a Unitree G1 humanoid, it can clear multiple obstacles in a row, adapt when obstacles get moved, and climb a wall up to 1.25m.

Rohan Paul

37,121 次观看 • 6 个月前

Depth Any Video with Scalable Synthetic Data AI physicists and chemists continue to make strides in depth estimation from video. Check out this new paper featuring some impressive examples. See the thread for more details (unfortunately no code yet). Abstract: Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse game environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates 0 - even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency.

Depth Any Video with Scalable Synthetic Data AI physicists and chemists continue to make strides in depth estimation from video. Check out this new paper featuring some impressive examples. See the thread for more details (unfortunately no code yet). Abstract: Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse game environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates 0 - even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency.

MrNeRF

27,428 次观看 • 1 年前

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷

Axis Robotics

28,096 次观看 • 8 个月前

Someone is vibe coding a live 3D news globe with Claude Code that pins real world events on a spinning Earth as they happen. Instead of doomscrolling a feed, you open a tab with a 3D Earth and watch the world light up in real time. Breaking news, conflicts, disasters and humanitarian alerts get pinned to the exact spot on the map, alongside live flights and upcoming rocket launches. Click any event and it pulls up a short brief with its actual sources, so you can see where it came from. People are already saying it's the only way they want to read news now, the kind of thing you keep open in a tab and just glance at all day.

Someone is vibe coding a live 3D news globe with Claude Code that pins real world events on a spinning Earth as they happen. Instead of doomscrolling a feed, you open a tab with a 3D Earth and watch the world light up in real time. Breaking news, conflicts, disasters and humanitarian alerts get pinned to the exact spot on the map, alongside live flights and upcoming rocket launches. Click any event and it pulls up a short brief with its actual sources, so you can see where it came from. People are already saying it's the only way they want to read news now, the kind of thing you keep open in a tab and just glance at all day.

0xMarioNawfal

39,104 次观看 • 2 个月前

This reported breakthrough apparently used a dataset that’s open for researchers. It’s the work of Eddy Xu, a teenager who was one of the first to get in on the video training data gold rush. He dropped out of Columbia last year to launch Build AI, which has raised around $22 million so far. Build AI’s Egocentric-1M dataset reportedly includes 1 million hours of data recorded using the startup’s self-developed devices across factories in Southeast Asia. A lot of it is from India. Xu has said he’s moved his team to Bengaluru, dedicating $10 million to get data from Indian factories. India has become one of the prime locations for collecting this kind of data. While enrolled at Columbia Engineering, Xu went viral in January 2025 after showing Meta Ray-Ban smart glasseshe modified to cheat at chess. The student, then 17, connected the device’s camera to a chess engine that calculated the best move and relayed it in real-time. The tech reached a wider audience thanks to popular streamer and chess master Alex Botez publicly tested them. Before college, the Long Island-raised Xu won DECA’s global business championship and sold an edtech startup that reached more than 178,000 users in 90 days. He also launched a startup called Omega Robotics in middle school, raising about $120,000 to run an independent, coach-free competitive robotics team out of a basement. Xu and co-founder Jonathan Jia, who serves as CTO, moved to San Francisco to build the first recording devices with a small team. They quickly moved operations to Shenzhen to quickly iterate and scale production. Build previously offered smaller datasets with 10,000 and 100,000 on Hugging Face but the 1M dataset requires emailing Xu directly. I’m sure he’s flooded with requests now.

This reported breakthrough apparently used a dataset that’s open for researchers. It’s the work of Eddy Xu, a teenager who was one of the first to get in on the video training data gold rush. He dropped out of Columbia last year to launch Build AI, which has raised around $22 million so far. Build AI’s Egocentric-1M dataset reportedly includes 1 million hours of data recorded using the startup’s self-developed devices across factories in Southeast Asia. A lot of it is from India. Xu has said he’s moved his team to Bengaluru, dedicating $10 million to get data from Indian factories. India has become one of the prime locations for collecting this kind of data. While enrolled at Columbia Engineering, Xu went viral in January 2025 after showing Meta Ray-Ban smart glasseshe modified to cheat at chess. The student, then 17, connected the device’s camera to a chess engine that calculated the best move and relayed it in real-time. The tech reached a wider audience thanks to popular streamer and chess master Alex Botez publicly tested them. Before college, the Long Island-raised Xu won DECA’s global business championship and sold an edtech startup that reached more than 178,000 users in 90 days. He also launched a startup called Omega Robotics in middle school, raising about $120,000 to run an independent, coach-free competitive robotics team out of a basement. Xu and co-founder Jonathan Jia, who serves as CTO, moved to San Francisco to build the first recording devices with a small team. They quickly moved operations to Shenzhen to quickly iterate and scale production. Build previously offered smaller datasets with 10,000 and 100,000 on Hugging Face but the 1M dataset requires emailing Xu directly. I’m sure he’s flooded with requests now.

Mike Kalil

13,025 次观看 • 1 个月前

𝗘𝘃𝗲𝗿𝘆𝗼𝗻𝗲’𝘀 𝘁𝗮𝗹𝗸𝗶𝗻𝗴 𝗮𝗯𝗼𝘂𝘁 “𝗣𝗵𝘆𝘀𝗶𝗰𝗮𝗹 𝗔𝗜" - the idea that we can simulate real-world environments so well that robots trained in simulation will work perfectly in reality. 𝗧𝗵𝗲 𝗽𝗿𝗼𝗺𝗶𝘀𝗲: Train in virtual worlds → deploy anywhere. 𝗧𝗵𝗲 𝗿𝗲𝗮𝗹𝗶𝘁𝘆: I’ve seen too many teams fall into this trap. After working with manipulation teams at Berkeley, Imperial, and Dyson, here’s the pattern: • 𝗪𝗲𝗲𝗸 𝟭: “Our policy works perfectly in simulation!” • 𝗪𝗲𝗲𝗸 𝟰: “Why doesn’t this work on real objects?” • 𝗠𝗼𝗻𝘁𝗵 𝟮: “We basically need to retrain from scratch with real data.” 𝗧𝗵𝗲 𝗴𝗮𝗽 𝘀𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻𝘀 𝗰𝗮𝗻’𝘁 𝗯𝗿𝗶𝗱𝗴𝗲: Unlike blind locomotion policies that can get away with sim-to-real transfer because they rely mainly on proprioception and contact forces, 𝘃𝗶𝘀𝗶𝗼𝗻-𝗴𝘂𝗶𝗱𝗲𝗱 𝗺𝗮𝗻𝗶𝗽𝘂𝗹𝗮𝘁𝗶𝗼𝗻 𝗶𝘀 𝗲𝘅𝘁𝗿𝗲𝗺𝗲𝗹𝘆 𝘀𝗲𝗻𝘀𝗶𝘁𝗶𝘃𝗲 𝘁𝗼 𝘃𝗶𝘀𝘂𝗮𝗹 𝗱𝗼𝗺𝗮𝗶𝗻 𝗴𝗮𝗽𝘀. • Real friction vs simulated surface textures • Manufacturing tolerances vs perfect CAD models • Dynamic lighting vs controlled virtual environments • Sensor noise vs instantaneous virtual readings 𝗛𝗲𝗿𝗲'𝘀 𝘄𝗵𝗮𝘁 𝗽𝗲𝗼𝗽𝗹𝗲 𝗱𝗼𝗻'𝘁 𝘁𝗮𝗹𝗸 𝗮𝗯𝗼𝘂𝘁: Building these detailed simulated environments takes forever. If it takes 7 days to build a simulated kitchen in simulation, wouldn't it be better to just collect real-world data in a real kitchen instead? 𝗗𝗼𝗻'𝘁 𝗴𝗲𝘁 𝗺𝗲 𝘄𝗿𝗼𝗻𝗴 - simulation is incredible for debugging, safety testing, and exploring edge cases. But it's not a magic solution to real-world deployment. 𝗪𝗵𝗮𝘁 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘄𝗼𝗿𝗸𝘀: Use simulation strategically while making real-world data collection as efficient and flexible as possible. This is why Neuracore focuses on streamlined real-world data infrastructure. Because no amount of virtual training can replace understanding how your robot actually behaves in actual environments. 𝗧𝗵𝗲 𝗽𝗵𝘆𝘀𝗶𝗰𝘀 𝗼𝗳 𝘆𝗼𝘂𝗿 𝗱𝗲𝗽𝗹𝗼𝘆𝗺𝗲𝗻𝘁 𝗲𝗻𝘃𝗶𝗿𝗼𝗻𝗺𝗲𝗻𝘁 𝗰𝗮𝗻'𝘁 𝗯𝗲 𝘀𝗶𝗺𝘂𝗹𝗮𝘁𝗲𝗱 𝗮𝘄𝗮𝘆. What’s been your experience with sim-to-real transfer?

𝗘𝘃𝗲𝗿𝘆𝗼𝗻𝗲’𝘀 𝘁𝗮𝗹𝗸𝗶𝗻𝗴 𝗮𝗯𝗼𝘂𝘁 “𝗣𝗵𝘆𝘀𝗶𝗰𝗮𝗹 𝗔𝗜" - the idea that we can simulate real-world environments so well that robots trained in simulation will work perfectly in reality. 𝗧𝗵𝗲 𝗽𝗿𝗼𝗺𝗶𝘀𝗲: Train in virtual worlds → deploy anywhere. 𝗧𝗵𝗲 𝗿𝗲𝗮𝗹𝗶𝘁𝘆: I’ve seen too many teams fall into this trap. After working with manipulation teams at Berkeley, Imperial, and Dyson, here’s the pattern: • 𝗪𝗲𝗲𝗸 𝟭: “Our policy works perfectly in simulation!” • 𝗪𝗲𝗲𝗸 𝟰: “Why doesn’t this work on real objects?” • 𝗠𝗼𝗻𝘁𝗵 𝟮: “We basically need to retrain from scratch with real data.” 𝗧𝗵𝗲 𝗴𝗮𝗽 𝘀𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻𝘀 𝗰𝗮𝗻’𝘁 𝗯𝗿𝗶𝗱𝗴𝗲: Unlike blind locomotion policies that can get away with sim-to-real transfer because they rely mainly on proprioception and contact forces, 𝘃𝗶𝘀𝗶𝗼𝗻-𝗴𝘂𝗶𝗱𝗲𝗱 𝗺𝗮𝗻𝗶𝗽𝘂𝗹𝗮𝘁𝗶𝗼𝗻 𝗶𝘀 𝗲𝘅𝘁𝗿𝗲𝗺𝗲𝗹𝘆 𝘀𝗲𝗻𝘀𝗶𝘁𝗶𝘃𝗲 𝘁𝗼 𝘃𝗶𝘀𝘂𝗮𝗹 𝗱𝗼𝗺𝗮𝗶𝗻 𝗴𝗮𝗽𝘀. • Real friction vs simulated surface textures • Manufacturing tolerances vs perfect CAD models • Dynamic lighting vs controlled virtual environments • Sensor noise vs instantaneous virtual readings 𝗛𝗲𝗿𝗲'𝘀 𝘄𝗵𝗮𝘁 𝗽𝗲𝗼𝗽𝗹𝗲 𝗱𝗼𝗻'𝘁 𝘁𝗮𝗹𝗸 𝗮𝗯𝗼𝘂𝘁: Building these detailed simulated environments takes forever. If it takes 7 days to build a simulated kitchen in simulation, wouldn't it be better to just collect real-world data in a real kitchen instead? 𝗗𝗼𝗻'𝘁 𝗴𝗲𝘁 𝗺𝗲 𝘄𝗿𝗼𝗻𝗴 - simulation is incredible for debugging, safety testing, and exploring edge cases. But it's not a magic solution to real-world deployment. 𝗪𝗵𝗮𝘁 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘄𝗼𝗿𝗸𝘀: Use simulation strategically while making real-world data collection as efficient and flexible as possible. This is why Neuracore focuses on streamlined real-world data infrastructure. Because no amount of virtual training can replace understanding how your robot actually behaves in actual environments. 𝗧𝗵𝗲 𝗽𝗵𝘆𝘀𝗶𝗰𝘀 𝗼𝗳 𝘆𝗼𝘂𝗿 𝗱𝗲𝗽𝗹𝗼𝘆𝗺𝗲𝗻𝘁 𝗲𝗻𝘃𝗶𝗿𝗼𝗻𝗺𝗲𝗻𝘁 𝗰𝗮𝗻'𝘁 𝗯𝗲 𝘀𝗶𝗺𝘂𝗹𝗮𝘁𝗲𝗱 𝗮𝘄𝗮𝘆. What’s been your experience with sim-to-real transfer?

Stephen James

25,347 次观看 • 11 个月前

🚨 THE BIGGEST BOTTLENECK IN AI ISN'T COMPUTING POWER ANYMORE IT'S MOVING DATA. Instead of laying new cables, Chinese researchers have upgraded existing fiber infrastructure by doing two things at once: Using three wavelength bands (C + L + S) instead of the usual two. Using four cores inside each fiber instead of one. Each core acts like an independent highway, and each band acts like an extra lane on that highway. Together, they’ve reportedly increased transmission capacity per core by nearly 50% and overall data throughput by up to 5×. This matters enormously for AI. Modern AI clusters move terabits of data per second between thousands of GPUs. The biggest bottleneck is often not the chips themselves, but moving data fast enough between them. If you can push 5× more data through the same physical cables, you can train bigger models faster and reduce network congestion. Why this is significant: • It shows multi-core + extended spectrum technology moving from labs into real-world commercial use • The system has already run over 35 km of existing telecom network • It could be especially useful for submarine cables and large-scale data center interconnects • China is also eyeing it for its “Eastern Data, Western Computing” project The deeper implication: We’re reaching the physical limits of how much data we can push through single-core fibers using traditional methods. By combining spatial multiplexing (multiple cores) with spectral multiplexing (more wavelength bands), engineers are finding new ways to keep scaling bandwidth without having to dig up the planet to lay new cables. This kind of breakthrough is quiet but foundational it’s the kind of infrastructure upgrade that will determine how fast AI and cloud computing can actually grow in the coming years. The future of data movement might not require more cables. It might just require smarter ones. How important do you think multi-core and multi-band fiber will be for keeping up with AI’s exploding data demands? Follow for more frontier networking, photonics, and infrastructure technology.

🚨 THE BIGGEST BOTTLENECK IN AI ISN'T COMPUTING POWER ANYMORE IT'S MOVING DATA. Instead of laying new cables, Chinese researchers have upgraded existing fiber infrastructure by doing two things at once: Using three wavelength bands (C + L + S) instead of the usual two. Using four cores inside each fiber instead of one. Each core acts like an independent highway, and each band acts like an extra lane on that highway. Together, they’ve reportedly increased transmission capacity per core by nearly 50% and overall data throughput by up to 5×. This matters enormously for AI. Modern AI clusters move terabits of data per second between thousands of GPUs. The biggest bottleneck is often not the chips themselves, but moving data fast enough between them. If you can push 5× more data through the same physical cables, you can train bigger models faster and reduce network congestion. Why this is significant: • It shows multi-core + extended spectrum technology moving from labs into real-world commercial use • The system has already run over 35 km of existing telecom network • It could be especially useful for submarine cables and large-scale data center interconnects • China is also eyeing it for its “Eastern Data, Western Computing” project The deeper implication: We’re reaching the physical limits of how much data we can push through single-core fibers using traditional methods. By combining spatial multiplexing (multiple cores) with spectral multiplexing (more wavelength bands), engineers are finding new ways to keep scaling bandwidth without having to dig up the planet to lay new cables. This kind of breakthrough is quiet but foundational it’s the kind of infrastructure upgrade that will determine how fast AI and cloud computing can actually grow in the coming years. The future of data movement might not require more cables. It might just require smarter ones. How important do you think multi-core and multi-band fiber will be for keeping up with AI’s exploding data demands? Follow for more frontier networking, photonics, and infrastructure technology.

TheNewPhysics

20,485 次观看 • 3 个月前