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.

Very excited to share our first public release after I joined Robbyant! We present Lingbot-Depth 👀 — a state-of-the-art depth foundation model trained with RGB-D MAE on millions of real & simulated RGBD pairs. 🔹 Camera depths as natural masks for RGB-D MAE modeling 🔹 Large-scale real + sim... show more

Yinghao Xu

3,910 subscribers

22,787 Aufrufe • vor 8 Monaten •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

10 Kommentare

Profilbild von Yinghao Xu

Yinghao Xuvor 8 Monaten

-Website: -Code: -Tech Report: -HuggingFace: Stay tuned!

Profilbild von David

Davidvor 8 Monaten

@robbyant_brain great work!

Profilbild von Ze Liu

Ze Liuvor 7 Monaten

@robbyant_brain great work!

Profilbild von Xiatao Sun

Xiatao Sunvor 7 Monaten

@robbyant_brain Really exciting release! RGB‑D MAE at scale with “sensor depth as masks” feels like a strong recipe for a depth foundation model, and the gains on transparent/reflective + thin structures are exactly the pain points for robotics.

Profilbild von Zwecharki of Immense AGI fundamentalism

Zwecharki of Immense AGI fundamentalismvor 7 Monaten

@robbyant_brain @grok does this apply to SfM for camera finding and such for gaussian splats?

Profilbild von Reza Sayar

Reza Sayarvor 7 Monaten

@robbyant_brain 👏👏👏👏👏👏😻

Profilbild von David Branca

David Brancavor 7 Monaten

@robbyant_brain Do you have any info on performance? Latency, etc

Profilbild von Philip McBride

Philip McBridevor 8 Monaten

@robbyant_brain Great work!

Profilbild von Nick Landolfi

Nick Landolfivor 8 Monaten

@robbyant_brain cool!!

Profilbild von Junlin Chang

Junlin Changvor 7 Monaten

@robbyant_brain Hi! Very interesting work! I wanna know how to address flying point.

Ähnliche Videos

The robot’s “eyes” just received a big upgrade. LingBot-Depth 2.0, a depth-completion model with half the depth error just dropped. 12/16 benchmarks topped. Glass, mirrors, and transparent objects are so easy for us humans, but so hard for robots, because they do not behave like ordinary surfaces in a camera pipeline. A robot that misunderstands a balcony window or a table edge, will have a completely false planning inside a false world. Huge implecation. LingBot-Depth 2.0 takes an RGB image plus a broken depth map from a sensor and then outputs a cleaner depth map and a usable 3D point cloud. Numbers on LingBot-Depth 2.0 - Excels on glass, mirrors & transparent objects — where traditional depth cameras fail - Training data: 3M → 150M (50x scale-up) - 12 out of 16 first-place rankings on depth completion benchmarks - RMSE cut in half: 0.132 → 0.062 on the hardest indoor scenes LingBot-Vision trained on boundaries, because object edges carry the geometry robots need. No human boundary labels are used, which makes this approach easier to scale. The open-sourced LingBot-Vision is the general vision backbone, and LingBot-Depth 2.0 is the depth model built on it.

The robot’s “eyes” just received a big upgrade. LingBot-Depth 2.0, a depth-completion model with half the depth error just dropped. 12/16 benchmarks topped. Glass, mirrors, and transparent objects are so easy for us humans, but so hard for robots, because they do not behave like ordinary surfaces in a camera pipeline. A robot that misunderstands a balcony window or a table edge, will have a completely false planning inside a false world. Huge implecation. LingBot-Depth 2.0 takes an RGB image plus a broken depth map from a sensor and then outputs a cleaner depth map and a usable 3D point cloud. Numbers on LingBot-Depth 2.0 - Excels on glass, mirrors & transparent objects — where traditional depth cameras fail - Training data: 3M → 150M (50x scale-up) - 12 out of 16 first-place rankings on depth completion benchmarks - RMSE cut in half: 0.132 → 0.062 on the hardest indoor scenes LingBot-Vision trained on boundaries, because object edges carry the geometry robots need. No human boundary labels are used, which makes this approach easier to scale. The open-sourced LingBot-Vision is the general vision backbone, and LingBot-Depth 2.0 is the depth model built on it.

Rohan Paul

25,312 Aufrufe • vor 2 Monaten

In collaboration with Intel, our Depth Fusion showcases the power of our LDM3D diffusion model in generating 360° views from text prompts provided by the user. The LDM3D diffusion model generates a 2D RGB image and its corresponding relative depth map providing a complete RGBD representation corresponding to the text prompt. The LDM 3D model is a specialized version of the stable diffusion V 1.4 model that has been modified to fit both image and depth map data.The model was then fine tuned on a subset of the Laion400M data set - large scale image caption data set. The depth maps used to fine tune our model were generated by the DPTBeiT large 512 depth estimation model that provides highly accurate relative depth estimates for each pixel. We take the generated 2D RGB image and depth map and use them to compute a 360° projection using touchdesigner. Touchdesigner is a versatile platform that allows for the creation of immersive and interactive multimedia experiences. Our application harnesses the power of touchdesigner to bring the generated 360° views to life, providing users with a unique and engaging way to experience their text prompts, whether it’s a description of a tranquil forest, a noisy cityscape or a futuristic sci fi world. Our depth fusion can bring these concepts to life in a vivid and immersive detail. - Scottie Fox, VP Engineering Blockade Labs ScottieFox #AI #VR #3D #gamedev #stablediffusion

In collaboration with Intel, our Depth Fusion showcases the power of our LDM3D diffusion model in generating 360° views from text prompts provided by the user. The LDM3D diffusion model generates a 2D RGB image and its corresponding relative depth map providing a complete RGBD representation corresponding to the text prompt. The LDM 3D model is a specialized version of the stable diffusion V 1.4 model that has been modified to fit both image and depth map data.The model was then fine tuned on a subset of the Laion400M data set - large scale image caption data set. The depth maps used to fine tune our model were generated by the DPTBeiT large 512 depth estimation model that provides highly accurate relative depth estimates for each pixel. We take the generated 2D RGB image and depth map and use them to compute a 360° projection using touchdesigner. Touchdesigner is a versatile platform that allows for the creation of immersive and interactive multimedia experiences. Our application harnesses the power of touchdesigner to bring the generated 360° views to life, providing users with a unique and engaging way to experience their text prompts, whether it’s a description of a tranquil forest, a noisy cityscape or a futuristic sci fi world. Our depth fusion can bring these concepts to life in a vivid and immersive detail. - Scottie Fox, VP Engineering Blockade Labs ScottieFox #AI #VR #3D #gamedev #stablediffusion

Blockade Labs

11,458 Aufrufe • vor 3 Jahren

Jason Ma reveals the first true scaling law in robotics: robots trained on 1 million hours of first-person human video predictably get better without ever seeing robot data "Yesterday we announced our new flagship robot foundation model called Dyna-2. It's very significant for the entire field for many different reasons. First of all, it's the first robot foundation model trained on at least one million hours of data. That in itself is a very challenging infrastructure challenge." "What we demonstrated is that even if the one million hours of data is entirely first-person video of humans doing manipulation tasks, we actually see scaling law for transferring to robot embodiments that the model has never seen before." "By training on more human data, we actually see predictable performance improvement on robots. That's a big deal because the biggest challenge facing robot foundation models is that we don't have enough data." "Unlike language models, there are just not enough robotics data out there, and it's very difficult to collect these robot data. Only by passively observing humans doing things can we scale this kind of data, and this is the first time we showed there is a lot of improvement that can transfer." Jason Ma Dyna Robotics

Jason Ma reveals the first true scaling law in robotics: robots trained on 1 million hours of first-person human video predictably get better without ever seeing robot data "Yesterday we announced our new flagship robot foundation model called Dyna-2. It's very significant for the entire field for many different reasons. First of all, it's the first robot foundation model trained on at least one million hours of data. That in itself is a very challenging infrastructure challenge." "What we demonstrated is that even if the one million hours of data is entirely first-person video of humans doing manipulation tasks, we actually see scaling law for transferring to robot embodiments that the model has never seen before." "By training on more human data, we actually see predictable performance improvement on robots. That's a big deal because the biggest challenge facing robot foundation models is that we don't have enough data." "Unlike language models, there are just not enough robotics data out there, and it's very difficult to collect these robot data. Only by passively observing humans doing things can we scale this kind of data, and this is the first time we showed there is a lot of improvement that can transfer." Jason Ma Dyna Robotics

MTS

17,740 Aufrufe • vor 1 Monat

For the past 3 months, I've been heads down building and extending our multimodal data capture rig for Physical AI training. What you're seeing here is a single synchronized capture from our system — RGB, depth (stereo IR camera) feed; real-time 3D hand reconstruction, and a 23×20 tactile pressure grid from sensor gloves, all aligned and timestamped together. Getting here took a lot of unglamorous iteration. We reworked the pipeline to deliver clean, gap-free depth at scale after identifying early depth frame drops caused by data throughput bottlenecks in our capture box. We built per-channel calibration models that convert raw resistance signals from our flex sensor gloves into accurate, drift-corrected force values — solving the creep behavior that made a steady grip falsely appear to fade over time. Even though we’ve stayed quiet on socials, word of mouth spread — our rigs have already been deployed in data factories, manufacturing lines, and homes across the US, Europe, China, and Southeast Asia by our customers and ourselves. In the next few days, I'll be sharing more about how we're tackling rights-cleared data capture from the hardware level, and our closed-loop data-model flywheel — using model evaluation results to identify gaps and guide what data our customers need. More soon, stay tuned…

For the past 3 months, I've been heads down building and extending our multimodal data capture rig for Physical AI training. What you're seeing here is a single synchronized capture from our system — RGB, depth (stereo IR camera) feed; real-time 3D hand reconstruction, and a 23×20 tactile pressure grid from sensor gloves, all aligned and timestamped together. Getting here took a lot of unglamorous iteration. We reworked the pipeline to deliver clean, gap-free depth at scale after identifying early depth frame drops caused by data throughput bottlenecks in our capture box. We built per-channel calibration models that convert raw resistance signals from our flex sensor gloves into accurate, drift-corrected force values — solving the creep behavior that made a steady grip falsely appear to fade over time. Even though we’ve stayed quiet on socials, word of mouth spread — our rigs have already been deployed in data factories, manufacturing lines, and homes across the US, Europe, China, and Southeast Asia by our customers and ourselves. In the next few days, I'll be sharing more about how we're tackling rights-cleared data capture from the hardware level, and our closed-loop data-model flywheel — using model evaluation results to identify gaps and guide what data our customers need. More soon, stay tuned…

Grace Zhang

54,991 Aufrufe • vor 3 Monaten

Meta presents Sapiens Foundation for Human Vision Models discuss: We present Sapiens, a family of models for four fundamental human-centric vision tasks - 2D pose estimation, body-part segmentation, depth estimation, and surface normal prediction. Our models natively support 1K high-resolution inference and are extremely easy to adapt for individual tasks by simply fine-tuning models pretrained on over 300 million in-the-wild human images. We observe that, given the same computational budget, self-supervised pretraining on a curated dataset of human images significantly boosts the performance for a diverse set of human-centric tasks. The resulting models exhibit remarkable generalization to in-the-wild data, even when labeled data is scarce or entirely synthetic. Our simple model design also brings scalability - model performance across tasks improves as we scale the number of parameters from 0.3 to 2 billion. Sapiens consistently surpasses existing baselines across various human-centric benchmarks. We achieve significant improvements over the prior state-of-the-art on Humans-5K (pose) by 7.6 mAP, Humans-2K (part-seg) by 17.1 mIoU, Hi4D (depth) by 22.4% relative RMSE, and THuman2 (normal) by 53.5% relative angular error.

Meta presents Sapiens Foundation for Human Vision Models discuss: We present Sapiens, a family of models for four fundamental human-centric vision tasks - 2D pose estimation, body-part segmentation, depth estimation, and surface normal prediction. Our models natively support 1K high-resolution inference and are extremely easy to adapt for individual tasks by simply fine-tuning models pretrained on over 300 million in-the-wild human images. We observe that, given the same computational budget, self-supervised pretraining on a curated dataset of human images significantly boosts the performance for a diverse set of human-centric tasks. The resulting models exhibit remarkable generalization to in-the-wild data, even when labeled data is scarce or entirely synthetic. Our simple model design also brings scalability - model performance across tasks improves as we scale the number of parameters from 0.3 to 2 billion. Sapiens consistently surpasses existing baselines across various human-centric benchmarks. We achieve significant improvements over the prior state-of-the-art on Humans-5K (pose) by 7.6 mAP, Humans-2K (part-seg) by 17.1 mIoU, Hi4D (depth) by 22.4% relative RMSE, and THuman2 (normal) by 53.5% relative angular error.

AK

151,614 Aufrufe • vor 2 Jahren

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

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

72,778 Aufrufe • vor 2 Monaten

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 →

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

119,660 Aufrufe • vor 1 Monat

Prince of Persia – Community Project The Prince of Persia community has come together to create its own project following the cancellation of the Sands of Time remake! This is not a traditional remake, but a large-scale mod aiming to replace most 3D models with high-quality assets and introduce global illumination through path tracing technology. We’re working on an official website — stay tuned! Current Status: Early development⏳ About the Project: 🔹 Public and community-driven 🔹 Fully playable 🔹 Focused on next-gen visuals while preserving the original experience We are looking for contributors! 🔹 3D Modelers / Artists 🔹 Texture Artists 🔹 Programmers 🔹 RTX Remix experts This is a hobby project with no payment, focused on long-term development without strict deadlines. We’re looking for passionate people who enjoy creating and want to be part of something special✨ Join us on Discord and become part of the team: 👉🏻 Follow our progress, share feedback, or get involved in development!😉 #PrinceofPersia #ThePoPProject

Prince of Persia – Community Project The Prince of Persia community has come together to create its own project following the cancellation of the Sands of Time remake! This is not a traditional remake, but a large-scale mod aiming to replace most 3D models with high-quality assets and introduce global illumination through path tracing technology. We’re working on an official website — stay tuned! Current Status: Early development⏳ About the Project: 🔹 Public and community-driven 🔹 Fully playable 🔹 Focused on next-gen visuals while preserving the original experience We are looking for contributors! 🔹 3D Modelers / Artists 🔹 Texture Artists 🔹 Programmers 🔹 RTX Remix experts This is a hobby project with no payment, focused on long-term development without strict deadlines. We’re looking for passionate people who enjoy creating and want to be part of something special✨ Join us on Discord and become part of the team: 👉🏻 Follow our progress, share feedback, or get involved in development!😉 #PrinceofPersia #ThePoPProject

PoP Universe

67,703 Aufrufe • vor 5 Monaten

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models

AK

62,719 Aufrufe • vor 2 Jahren

Dear friends, I’m very excited to announce our next ambitious exhibition opening in Guggenheim Bilbao on the 7th of March! We are transforming my hero — Frank Gehry’s iconic architecture with 50 channel projectors into “Living Architecture.” With respect and homage to his body of work, our Large Architecture Model, trained on 35 million images of the world's architecture, will be dreaming of the future of architecture, urban, and future cities during the 6 months of our show in real-time using light and data as a pigment. I do believe Architecture is beyond concrete, steel and glass, and I’m deeply honored to share our work in an incredible museum and grateful to the Guggenheim Bilbao team for making our dream a reality!

Dear friends, I’m very excited to announce our next ambitious exhibition opening in Guggenheim Bilbao on the 7th of March! We are transforming my hero — Frank Gehry’s iconic architecture with 50 channel projectors into “Living Architecture.” With respect and homage to his body of work, our Large Architecture Model, trained on 35 million images of the world's architecture, will be dreaming of the future of architecture, urban, and future cities during the 6 months of our show in real-time using light and data as a pigment. I do believe Architecture is beyond concrete, steel and glass, and I’m deeply honored to share our work in an incredible museum and grateful to the Guggenheim Bilbao team for making our dream a reality!

Refik Anadol

15,562 Aufrufe • vor 1 Jahr

Today we are excited to open up Neuracore to the academic community! Neuracore is a new data foundation built to accelerate robot learning by removing one of the field’s biggest bottlenecks: capturing and working with high-fidelity multimodal robotics data. For the first time, researchers can store, view, and work with robotics data in a cloud-native system built specifically for large-scale learning, and we are making this core platform completely free for academia. The platform lets teams capture every sensor at its native rate, store and visualize data without loss, and then train and deploy models locally using our open-source code (Link in the comments). We are rolling out access to select academic institutions first. Anyone with an academic email can sign up, and if your institution is not part of the initial rollout, you will be able to join the waitlist directly. Beyond providing this infrastructure, we see an opportunity to build a global community where engineers and researchers can share, collaborate, and advance the frontier of robot learning together. Supported by our recent $3M pre-seed round led by Earlybird VC, we are excited to take this mission even further. Our long-term goal is for Neuracore to become the natural home for cutting-edge robot learning algorithms and real-world robotics experimentation, helping accelerate the next wave of Physical AI.

Today we are excited to open up Neuracore to the academic community! Neuracore is a new data foundation built to accelerate robot learning by removing one of the field’s biggest bottlenecks: capturing and working with high-fidelity multimodal robotics data. For the first time, researchers can store, view, and work with robotics data in a cloud-native system built specifically for large-scale learning, and we are making this core platform completely free for academia. The platform lets teams capture every sensor at its native rate, store and visualize data without loss, and then train and deploy models locally using our open-source code (Link in the comments). We are rolling out access to select academic institutions first. Anyone with an academic email can sign up, and if your institution is not part of the initial rollout, you will be able to join the waitlist directly. Beyond providing this infrastructure, we see an opportunity to build a global community where engineers and researchers can share, collaborate, and advance the frontier of robot learning together. Supported by our recent $3M pre-seed round led by Earlybird VC, we are excited to take this mission even further. Our long-term goal is for Neuracore to become the natural home for cutting-edge robot learning algorithms and real-world robotics experimentation, helping accelerate the next wave of Physical AI.

Neuracore

40,810 Aufrufe • vor 10 Monaten

We are excited to share our #CORL2024 paper on learning quadrotor obstacle avoidance from the visual stream of a single #eventcamera! Trained entirely in simulation! We demonstrate obstacle avoidance both in the dark and in a forest up to 5m/s. PDF: Video: Project page: Event cameras are sensors that output per-pixel-level intensity changes at microsecond latency resolution; they feature nearly zero motion blur and high dynamic range but produce a very large volume of events under significant ego-motion and further lack a high-fidelity continuous-time sensor model in simulation, making direct #sim2real transfer not possible. By leveraging depth prediction as a pretext task, we pre-train a reactive obstacle avoidance policy with “approximated” simulated events and then fine-tune the perception component with limited events-and-depth real-world data. This technique bridges the sim2real gap for #eventcameras! As at the current state, there is no continuous-time sensor model for event cameras, we hope that this work can finally spur future research leveraging simulation for training event-vision-based policies to create faster, agile robots! Kudos to Anish Bhattacharya, @marcocannic, Vijay Kumar Nikolai Matni UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) GRASP Laboratory Penn Engineering

We are excited to share our #CORL2024 paper on learning quadrotor obstacle avoidance from the visual stream of a single #eventcamera! Trained entirely in simulation! We demonstrate obstacle avoidance both in the dark and in a forest up to 5m/s. PDF: Video: Project page: Event cameras are sensors that output per-pixel-level intensity changes at microsecond latency resolution; they feature nearly zero motion blur and high dynamic range but produce a very large volume of events under significant ego-motion and further lack a high-fidelity continuous-time sensor model in simulation, making direct #sim2real transfer not possible. By leveraging depth prediction as a pretext task, we pre-train a reactive obstacle avoidance policy with “approximated” simulated events and then fine-tune the perception component with limited events-and-depth real-world data. This technique bridges the sim2real gap for #eventcameras! As at the current state, there is no continuous-time sensor model for event cameras, we hope that this work can finally spur future research leveraging simulation for training event-vision-based policies to create faster, agile robots! Kudos to Anish Bhattacharya, @marcocannic, Vijay Kumar Nikolai Matni UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) GRASP Laboratory Penn Engineering

Davide Scaramuzza

17,219 Aufrufe • vor 1 Jahr

Excited to share this demo from Over the Reality! Watch our Unitree Go2 robodog navigating our office while reconstructing the 3D space in real-time using the VGGT-based foundation vision model. This is a prime example of machine perception in action, turning raw RGB camera feeds into rich, detailed 3D maps! The robodog's RGB cam generates a dense, textured 3D reconstruction via VGGT from a few photograms, capturing nuances like object shapes and surfaces with impressive fidelity (main view). Compare that to the standard LiDAR system (top right), it's sparser, more point-cloud focused, lacking the visual richness. Vision models are closing the gap fast! What's powering this? VGGT, a cutting-edge foundation model for 3D perception, trained on datasets orders of magnitude smaller than our massive OVER 3D maps dataset. Imagine the leap when we apply OVER's scale to VGGT-like transformer based architectures, denser reconstructions, better generalization, revolutionary for robotics, machine perception & AR! Stay tuned for more breakthroughs at the intersection of AI, robotics, and DePIN. We're building the future of Physical AI and Spatial Computing at Over the Reality 🌐 What do you think, ready for robodogs in your world? Drop your thoughts! 🤖🌐

Excited to share this demo from Over the Reality! Watch our Unitree Go2 robodog navigating our office while reconstructing the 3D space in real-time using the VGGT-based foundation vision model. This is a prime example of machine perception in action, turning raw RGB camera feeds into rich, detailed 3D maps! The robodog's RGB cam generates a dense, textured 3D reconstruction via VGGT from a few photograms, capturing nuances like object shapes and surfaces with impressive fidelity (main view). Compare that to the standard LiDAR system (top right), it's sparser, more point-cloud focused, lacking the visual richness. Vision models are closing the gap fast! What's powering this? VGGT, a cutting-edge foundation model for 3D perception, trained on datasets orders of magnitude smaller than our massive OVER 3D maps dataset. Imagine the leap when we apply OVER's scale to VGGT-like transformer based architectures, denser reconstructions, better generalization, revolutionary for robotics, machine perception & AR! Stay tuned for more breakthroughs at the intersection of AI, robotics, and DePIN. We're building the future of Physical AI and Spatial Computing at Over the Reality 🌐 What do you think, ready for robodogs in your world? Drop your thoughts! 🤖🌐

Over the Reality 🌐

359,764 Aufrufe • vor 1 Jahr

Dear friends, l'm so excited to share our most recent and ambitious project: Large Nature Model / Living Art in collaboration with NVIDIA . This week, before the epic GTC keynote of Jensen Huang, I was able to transform 20 meters high and 40 meters wide 16k resolution media wall into real-time Al Data Sculpture. Artwork fed by real-time Large Nature Model's flora, fauna, and landscapes Al text to video outputs! Together with NVIDIA Al researchers, we finetuned Getty Image's foundational model with our 2.5 million ethically sourced Nature dataset and made a significant scale physical installation with Al scent and Al character as well. We are deeply grateful to the entire NVIDIA team for the collaboration, support, and pushing the boundaries of imagination in the age of Al!

Dear friends, l'm so excited to share our most recent and ambitious project: Large Nature Model / Living Art in collaboration with NVIDIA . This week, before the epic GTC keynote of Jensen Huang, I was able to transform 20 meters high and 40 meters wide 16k resolution media wall into real-time Al Data Sculpture. Artwork fed by real-time Large Nature Model's flora, fauna, and landscapes Al text to video outputs! Together with NVIDIA Al researchers, we finetuned Getty Image's foundational model with our 2.5 million ethically sourced Nature dataset and made a significant scale physical installation with Al scent and Al character as well. We are deeply grateful to the entire NVIDIA team for the collaboration, support, and pushing the boundaries of imagination in the age of Al!

Refik Anadol

51,355 Aufrufe • vor 2 Jahren

Drop 11/14: We have been releasing Sarvam’s models and products - and yes, there is more to come. But today, we are excited to share the real and diverse impact our work is driving at scale. First up, preserving our cultural heritage. We have been working with Ekatra Foundation and Navajivan Trust (founded by Mahatma Gandhi) to digitize Gujarati documents from the early 19th and 20th centuries. Through this collaboration, we built out a whole product - Sarvam Akshar. Powered by the Sarvam Vision model, Akshar delivers state-of-the-art accuracy enabling reliable digitization of complex, real-world documents. Akshar sets the stage for our effort to leverage foundational models to preserve India’s cultural heritage - a direction we will double down on. Read more in our blog:

Drop 11/14: We have been releasing Sarvam’s models and products - and yes, there is more to come. But today, we are excited to share the real and diverse impact our work is driving at scale. First up, preserving our cultural heritage. We have been working with Ekatra Foundation and Navajivan Trust (founded by Mahatma Gandhi) to digitize Gujarati documents from the early 19th and 20th centuries. Through this collaboration, we built out a whole product - Sarvam Akshar. Powered by the Sarvam Vision model, Akshar delivers state-of-the-art accuracy enabling reliable digitization of complex, real-world documents. Akshar sets the stage for our effort to leverage foundational models to preserve India’s cultural heritage - a direction we will double down on. Read more in our blog:

Pratyush Kumar

61,100 Aufrufe • vor 7 Monaten

A jaw-on-the-floor moment for the whole team was when we had taught the robot to pick up a baggie and shake a usb-brick out with the left hand. During one rollout, it decided to do it with the right hand—and we all stopped. We were stunned, because no one taught it to do that. This was one of the first models with our pretrained base, and it was the moment we realized there was something here that was very different from what any of us had ever built before. We now know this to be part of the improvisational intelligence that emerges with large-scale pretraining on real manipulation data, and it continues to be one of the things that we believe to be worth pushing forward as capability. Our CTO and cofounder Andy Barry joined Association for Advancing Automation to share more about these moments, and how his time at Boston Dynamics and the Broad Institute drove him to focus on building the 🧠 for robots at Generalist. Full ep 👇

A jaw-on-the-floor moment for the whole team was when we had taught the robot to pick up a baggie and shake a usb-brick out with the left hand. During one rollout, it decided to do it with the right hand—and we all stopped. We were stunned, because no one taught it to do that. This was one of the first models with our pretrained base, and it was the moment we realized there was something here that was very different from what any of us had ever built before. We now know this to be part of the improvisational intelligence that emerges with large-scale pretraining on real manipulation data, and it continues to be one of the things that we believe to be worth pushing forward as capability. Our CTO and cofounder Andy Barry joined Association for Advancing Automation to share more about these moments, and how his time at Boston Dynamics and the Broad Institute drove him to focus on building the 🧠 for robots at Generalist. Full ep 👇

Generalist

11,117 Aufrufe • vor 3 Monaten

In just one week, Binh and I trained a full-body Unitree G1. Here's a recap: 1. Secured a Unitree G1 humanoid through a LinkedIn post 2. Deployed TWIST2 full-body teleoperation pipelines 3. Adapted TWIST2 for Zed stereo camera & collected full-body teleoperation samples (carried by Binh ) 4. Adapted & fine-tuned NVIDIA Gr00T N1.5 VLA on the TWIST2 public datasets, which I fine-tuned on an 8xNVIDIA H100 Cluster. We picked Gr00T N1.5 as it was trained with Unitree G1 embodiment data. 5. Adapted the TWIST2 codebase to stream in the actions from Gr00T via ZMQ using a co-located NVIDIA H100 for ~200ms inference latency 6. Tested the model in sim, then deployed to the real-world Unitree G1. We streamed a training sample observation to the VLA (as we didn't want to break robot in case real observations were OOD) We were the first team in the world to deploy the full TWIST2 data collection pipeline to the unitree g1 :) Much more work ahead though, which I'll work on as a side-project over the next months: 1. Exploring the various types of 'world models': video backbones, dynamics models, v-jepa-2 models. I believe these will generalize better & train much more data-efficiently than VLM backbones 2. Speeding up inference - I believe low-latency robotics inference will be a big challenge. There are many works in video diffusion which I'd like to test (e.g. SageAttention, SparseAttention, Drifting Models). Perhaps also writing custom CUDA kernels. 3. Economics of inference scaling :) What will be the compute demands as we scale inference up to millions of humanoids? Will it run on edge or on distributed 'co-located' inference clusters? These are questions I'd like to answer. Adapted TWIST2 codebase: Adapted Gr00T-N1.5 codebase: The ETH Robotics Club are doing a cool GTC Golden ticket competition with NVIDIA , so this is my submission :) The DGX Spark compute will get me a long way with initial prototyping & especially working on inference optimization for next-gen Blackwell GPUs #NVIDIAGTC #GOLDENTICKET #ETHRC

In just one week, Binh and I trained a full-body Unitree G1. Here's a recap: 1. Secured a Unitree G1 humanoid through a LinkedIn post 2. Deployed TWIST2 full-body teleoperation pipelines 3. Adapted TWIST2 for Zed stereo camera & collected full-body teleoperation samples (carried by Binh ) 4. Adapted & fine-tuned NVIDIA Gr00T N1.5 VLA on the TWIST2 public datasets, which I fine-tuned on an 8xNVIDIA H100 Cluster. We picked Gr00T N1.5 as it was trained with Unitree G1 embodiment data. 5. Adapted the TWIST2 codebase to stream in the actions from Gr00T via ZMQ using a co-located NVIDIA H100 for ~200ms inference latency 6. Tested the model in sim, then deployed to the real-world Unitree G1. We streamed a training sample observation to the VLA (as we didn't want to break robot in case real observations were OOD) We were the first team in the world to deploy the full TWIST2 data collection pipeline to the unitree g1 :) Much more work ahead though, which I'll work on as a side-project over the next months: 1. Exploring the various types of 'world models': video backbones, dynamics models, v-jepa-2 models. I believe these will generalize better & train much more data-efficiently than VLM backbones 2. Speeding up inference - I believe low-latency robotics inference will be a big challenge. There are many works in video diffusion which I'd like to test (e.g. SageAttention, SparseAttention, Drifting Models). Perhaps also writing custom CUDA kernels. 3. Economics of inference scaling :) What will be the compute demands as we scale inference up to millions of humanoids? Will it run on edge or on distributed 'co-located' inference clusters? These are questions I'd like to answer. Adapted TWIST2 codebase: Adapted Gr00T-N1.5 codebase: The ETH Robotics Club are doing a cool GTC Golden ticket competition with NVIDIA , so this is my submission :) The DGX Spark compute will get me a long way with initial prototyping & especially working on inference optimization for next-gen Blackwell GPUs #NVIDIAGTC #GOLDENTICKET #ETHRC

Arnie Ramesh

23,236 Aufrufe • vor 7 Monaten

GS^3: Efficient Relighting with Triple Gaussian Splatting Abstract: We present a spatial and angular Gaussian based representation and a triple splatting process, for real-time, high-quality novel lighting-and-view synthesis from multi-view point-lit input images. To describe complex ap pearance, we employ a Lambertian plus a mixture of angular Gaussians as an effective reflectance function for each spatial Gaussian. To generate self-shadow, we splat all spatial Gaussians towards the light source to obtain shadow values, which are further refined by a small multi-layer perceptron. To compensate for other effects like global illumination, another network is trained to compute and add a per-spatial-Gaussian RGB tuple. The effectiveness of our representation is demonstrated on 30 samples with a wide variation in geometry (from solid to fluffy) and appearance (from translucent to anisotropic), as well as using different forms of input data, including rendered images of synthetic/reconstructed objects, photographs captured with a handheld camera and a flash, or from a professional lightstage. We achieve a training time of 40-70 minutes and a rendering speed of 90 fps on a single commodity GPU. Our results compare favorably with state-of-the-art techniques in terms of quality/performance.

GS^3: Efficient Relighting with Triple Gaussian Splatting Abstract: We present a spatial and angular Gaussian based representation and a triple splatting process, for real-time, high-quality novel lighting-and-view synthesis from multi-view point-lit input images. To describe complex ap pearance, we employ a Lambertian plus a mixture of angular Gaussians as an effective reflectance function for each spatial Gaussian. To generate self-shadow, we splat all spatial Gaussians towards the light source to obtain shadow values, which are further refined by a small multi-layer perceptron. To compensate for other effects like global illumination, another network is trained to compute and add a per-spatial-Gaussian RGB tuple. The effectiveness of our representation is demonstrated on 30 samples with a wide variation in geometry (from solid to fluffy) and appearance (from translucent to anisotropic), as well as using different forms of input data, including rendered images of synthetic/reconstructed objects, photographs captured with a handheld camera and a flash, or from a professional lightstage. We achieve a training time of 40-70 minutes and a rendering speed of 90 fps on a single commodity GPU. Our results compare favorably with state-of-the-art techniques in terms of quality/performance.

MrNeRF

17,821 Aufrufe • vor 2 Jahren