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。

What would a World Model look like if we start from a real embodied agent acting in the real world? It has to have: 1) A real, physically grounded and complex action space—not just abstract control signals. 2) Diverse, real-life scenarios and activities. Or in short: It has to... be annoyingly complex—in both the action and vision space—to even get close to real life. We did an initial attempt: Whole-Body Conditioned Egocentric Video Prediction. In collaboration with @dans_t123 , Amir Bar, Yann LeCun , trevordarrell and Jitendra MALIK. (For more details, check: What we did is very simple: Predict Egocentric Video from human Actions (PEVA) - Given the past video and a future action represented by relative 3D body pose, PEVA predicts how the world looks next—from the first-person view. By conditioning on kinematic pose trajectories, structured by the joint hierarchy of the body, it learns how physical actions shape perception.show more

Yutong Bai

4,929 subscribers

179,017 次观看 • 1 年前 •via X (Twitter)

健康养生科学技术教育

Anya Rossi• Live Now

Private livecam show

10 条评论

Yutong Bai 的头像

Yutong Bai1 年前

Let’s see some pixels! We mainly showcase three types of results - atom actions, long video rollouts and counterfactuals for planning.

Yutong Bai 的头像

Yutong Bai1 年前

For more atom actions:

Yutong Bai 的头像

Yutong Bai1 年前

For more long video rolling out examples

Yutong Bai 的头像

Yutong Bai1 年前

For counterfactual planning:

Yutong Bai 的头像

Yutong Bai1 年前

Here’s the project page with more results, failures and discussions we have, check these out!

Nova Circuit 的头像

Nova Circuit1 年前

@ylecun @JitendraMalikCV @trevordarrell @_amirbar @dans_t123 Integrating the complex action space with grounded physicality could yield more robust predictions compared to existing abstract models.

Yutong Bai 的头像

Yutong Bai1 年前

Exactly — from a first-person human perspective, even something as simple as walking causes the first-person video to shake along with the person’s gait and movement. So if the model can learn the relationship between this motion and the video, it could potentially be more robust to that shaking when understanding the goal.

Max Profiton 的头像

Max Profiton1 年前

@ylecun @JitendraMalikCV @trevordarrell @_amirbar @dans_t123 Integrating embodiment adds significant modeling subtleties. Promising direction.

Animesh Garg 的头像

Animesh Garg1 年前

this is very interesting and timely @GnosisYu did some work on egocentric camera control recently at ICLR -- EgoSim wondering what comparisons may look like as well as what architectural changes were needed for precise camera control. @ylecun, @trevordarrell, @JitendraMalikCV

Yutong Bai 的头像

Yutong Bai1 年前

@ylecun @JitendraMalikCV @trevordarrell @_amirbar @dans_t123 @GnosisYu Thank you for pointing this out! We will cite this in the next version. Yeah camera pose is very important. I think human’s complex body motion make this problem even harder.

相关视频

🦾 From seeing to doing. We're closing the loop between video prediction and real-world action. On the final day of Robbyant Open Source Week, we bring you LingBot-VA—the world's first causal video-action world model for generalist robot control. 🔥 Key Highlights: 🤖 Predicts & Acts: A single model generates both future video and the actions to make it real. 🧠 Remembers the Past: True long-term memory for complex, sequential tasks. ⚡ Learns in a Snap: Masters new skills with just 30-50 real-world demos. The result? 📈 SOTA on RoboTwin (92.9%) and LIBERO (98.5%), +20% over π0.5 on challenging real-world long-horizon & high-precision tasks. More below 👇 #AI #Robotics #EmbodiedAI #WorldModel #OpenSource

🦾 From seeing to doing. We're closing the loop between video prediction and real-world action. On the final day of Robbyant Open Source Week, we bring you LingBot-VA—the world's first causal video-action world model for generalist robot control. 🔥 Key Highlights: 🤖 Predicts & Acts: A single model generates both future video and the actions to make it real. 🧠 Remembers the Past: True long-term memory for complex, sequential tasks. ⚡ Learns in a Snap: Masters new skills with just 30-50 real-world demos. The result? 📈 SOTA on RoboTwin (92.9%) and LIBERO (98.5%), +20% over π0.5 on challenging real-world long-horizon & high-precision tasks. More below 👇 #AI #Robotics #EmbodiedAI #WorldModel #OpenSource

Robbyant

704,157 次观看 • 6 个月前

Yann LeCun (Yann LeCun) just revealed why he left Meta, Why are LLMs an extremely narrow field, and how "World model" is the way to build really meaningful agentic future. 🎯 Beautiful and simple in 1.5 mints. "I can’t imagine we can build agentic systems without those systems having the ability to predict, in advance, what the consequences of their actions are going to be. The way we act in the world is that we can predict the consequences of our actions, and that’s what allows us to plan. So what is a world model? Given the state of an environment, a system you want to control at time t, and given an action or intervention you imagine taking, can you predict the state of the world (or the system) at time t + 1? If you can, that’s a world model. You don’t do this at a pixel level, if it’s video. You do this in an abstract representation space, and that’s a crucial key insight." --- From 'AI House Davos" YT channel (full link in comment)

Yann LeCun (Yann LeCun) just revealed why he left Meta, Why are LLMs an extremely narrow field, and how "World model" is the way to build really meaningful agentic future. 🎯 Beautiful and simple in 1.5 mints. "I can’t imagine we can build agentic systems without those systems having the ability to predict, in advance, what the consequences of their actions are going to be. The way we act in the world is that we can predict the consequences of our actions, and that’s what allows us to plan. So what is a world model? Given the state of an environment, a system you want to control at time t, and given an action or intervention you imagine taking, can you predict the state of the world (or the system) at time t + 1? If you can, that’s a world model. You don’t do this at a pixel level, if it’s video. You do this in an abstract representation space, and that’s a crucial key insight." --- From 'AI House Davos" YT channel (full link in comment)

Rohan Paul

383,490 次观看 • 6 个月前

Diffusion has shown great promise for generating robot **actions**, can it act as a **world model** to generate the future conditioned on actions? In our work led by han qi Haocheng Yin and in collaboration with Yilun Du, we show a **controllable** action-conditioned video diffusion model can produce photorealistic and (near) physics-accurate future predictions. This ability strengthens the policy via: - ranking different action proposals and selecting the best, or - **visual** trajectory optimization by optimizing the action proposals using gradient ascent. Learn more about Generative Predictive Control (GPC) at:

Diffusion has shown great promise for generating robot actions, can it act as a world model to generate the future conditioned on actions? In our work led by han qi Haocheng Yin and in collaboration with Yilun Du, we show a controllable action-conditioned video diffusion model can produce photorealistic and (near) physics-accurate future predictions. This ability strengthens the policy via: - ranking different action proposals and selecting the best, or - visual trajectory optimization by optimizing the action proposals using gradient ascent. Learn more about Generative Predictive Control (GPC) at:

Heng Yang

38,434 次观看 • 1 年前

Yann LeCun (Yann LeCun ) beautifully explains how the architecture and principles used to train LLMs can not be extended to teach AI the real-world intelligence. In 1 line: LLMs excel where intelligence equals sequence prediction over symbols. Real-world intelligence requires learned world models, abstraction, causality, and action planning under uncertainty, which current next-token training does not provide. He says current LLMs learn by predicting the next token. That objective works very well when the task itself can be reduced to manipulating discrete symbols and sequences. Math, physics problem solving on paper, and coding fit this pattern because success largely comes from searching and composing the right sequences of symbols, equations, or program tokens. With enough data and scale, these models get very good at that kind of structured sequence prediction. Real-world intelligence is different. The physical world is continuous, noisy, uncertain, and high dimensional. To act in it, a system needs internal models that capture objects, dynamics, causality, constraints from the body, and the outcomes of actions over time. Humans and animals build abstract representations from rich sensory streams, then make predictions in that abstract space, not at the raw pixel level. That is why a child can learn intuitive physics, plan multi-step actions, and adapt quickly in new situations with little data. His claim about saturation follows from this gap. Scaling token prediction keeps improving symbol manipulation tasks like math and code, but it hits limits on embodied reasoning and common sense because text alone does not provide the right learning signals for world models. Predicting the next word cannot efficiently teach contact forces, affordances, occlusion, friction, or how actions change the state of the environment. For that, he argues we need architectures that learn abstractions from sensory data and predict futures in abstract latent spaces, then use those predictions to plan actions toward goals with built-in guardrails. --- From 'Pioneer Works' YT Channel (link in comment)

Yann LeCun (Yann LeCun ) beautifully explains how the architecture and principles used to train LLMs can not be extended to teach AI the real-world intelligence. In 1 line: LLMs excel where intelligence equals sequence prediction over symbols. Real-world intelligence requires learned world models, abstraction, causality, and action planning under uncertainty, which current next-token training does not provide. He says current LLMs learn by predicting the next token. That objective works very well when the task itself can be reduced to manipulating discrete symbols and sequences. Math, physics problem solving on paper, and coding fit this pattern because success largely comes from searching and composing the right sequences of symbols, equations, or program tokens. With enough data and scale, these models get very good at that kind of structured sequence prediction. Real-world intelligence is different. The physical world is continuous, noisy, uncertain, and high dimensional. To act in it, a system needs internal models that capture objects, dynamics, causality, constraints from the body, and the outcomes of actions over time. Humans and animals build abstract representations from rich sensory streams, then make predictions in that abstract space, not at the raw pixel level. That is why a child can learn intuitive physics, plan multi-step actions, and adapt quickly in new situations with little data. His claim about saturation follows from this gap. Scaling token prediction keeps improving symbol manipulation tasks like math and code, but it hits limits on embodied reasoning and common sense because text alone does not provide the right learning signals for world models. Predicting the next word cannot efficiently teach contact forces, affordances, occlusion, friction, or how actions change the state of the environment. For that, he argues we need architectures that learn abstractions from sensory data and predict futures in abstract latent spaces, then use those predictions to plan actions toward goals with built-in guardrails. --- From 'Pioneer Works' YT Channel (link in comment)

Rohan Paul

104,460 次观看 • 7 个月前

Meet VLA-JEPA, the first world model policy in LeRobot 🌍 Most VLAs just learn observation → action. VLA-JEPA also learns action-relevant dynamics: during training, a JEPA world model has to predict upcoming frames in latent space from the model's own actions. Then, at inference, the world model disappears entirely, leaving a standard, fast VLA: Qwen backbone, action head, nothing extra. In our demo it was fine-tuned on just 13 examples and ran in real time on an NVIDIA DGX Spark. On a consumer RTX 3080, it hits 10Hz using under 6GB of VRAM. World-model supervision during training, zero extra cost at inference. That's the whole trick.

Meet VLA-JEPA, the first world model policy in LeRobot 🌍 Most VLAs just learn observation → action. VLA-JEPA also learns action-relevant dynamics: during training, a JEPA world model has to predict upcoming frames in latent space from the model's own actions. Then, at inference, the world model disappears entirely, leaving a standard, fast VLA: Qwen backbone, action head, nothing extra. In our demo it was fine-tuned on just 13 examples and ran in real time on an NVIDIA DGX Spark. On a consumer RTX 3080, it hits 10Hz using under 6GB of VRAM. World-model supervision during training, zero extra cost at inference. That's the whole trick.

LeRobot

42,109 次观看 • 25 天前

Robot policies must be both reliable and highly capable to be useful; the best way to achieve this level of performance is with reinforcement learning. However, for reinforcement learning you are usually stuck between two difficult options: reinforcement in the real world is often risky and expensive, while reinforcement learning in a traditional simulator takes a lot of engineering work and has a persistent sim-to-real gap. What if instead you could train your robot purely in a world model? RISE by Jiazhi Yang et al. uses a compositional world model to predict the future and evaluate progress. This allows for a self-improving pipeline, which learns a world model from real data and then learns how the robot should perform different tasks. This pipeline results in a data-driven way to improve policy performance from real data but without real-world reinforcement learning. Watch Episode #86 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

Robot policies must be both reliable and highly capable to be useful; the best way to achieve this level of performance is with reinforcement learning. However, for reinforcement learning you are usually stuck between two difficult options: reinforcement in the real world is often risky and expensive, while reinforcement learning in a traditional simulator takes a lot of engineering work and has a persistent sim-to-real gap. What if instead you could train your robot purely in a world model? RISE by Jiazhi Yang et al. uses a compositional world model to predict the future and evaluate progress. This allows for a self-improving pipeline, which learns a world model from real data and then learns how the robot should perform different tasks. This pipeline results in a data-driven way to improve policy performance from real data but without real-world reinforcement learning. Watch Episode #86 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

RoboPapers

38,334 次观看 • 1 个月前

This is a video of world leaders/presidents singing "We are the world" by Michael Jackson Led by Trump Then Putin, Kim, Xi and other world presidents joined in It isn't real of course, but it's extremely thoughtful and it would be world-class humbling if this happened in real life! 💯 A part of me really feels the world needs something like this. Honestly

This is a video of world leaders/presidents singing "We are the world" by Michael Jackson Led by Trump Then Putin, Kim, Xi and other world presidents joined in It isn't real of course, but it's extremely thoughtful and it would be world-class humbling if this happened in real life! 💯 A part of me really feels the world needs something like this. Honestly

Vincent The Therapist

14,545 次观看 • 3 个月前

Breaking news: Cosmos 3 is here. They are attempting to do something completely new 🤯 Why is Physical AI much harder than building a chatbot? Understanding the world is not enough, robots need to predict it and act inside it. That's the idea behind NVIDIA Cosmos 3: → Reasoning model understands what's happening from video, images, text, and actions. → World model generates future states of the environment. → Action model generates the actions needed to achieve a goal. Previous systems often stitched these capabilities together using separate models. Cosmos 3 combines them into a single architecture with two components: ▪️ Reasoner Tower Analyzes observations and builds an understanding of objects, motion, interactions, and physical context. ▪️ Generator Tower Uses that understanding to generate future videos and action sequences that obey physical constraints. So Cosmos 3 moves from: Perception → Model A Prediction → Model B Actions → Model C to: Perception + Prediction + Actions → One unified system. The goal is to make robots, autonomous vehicles, and smart environments better at answering three questions: 1. What is happening? 2. What will happen next? 3. What should I do? That's a big shift from today's AI, which mostly focuses on generating text. And check the benchmarks! Physical AI needs to generate decisions that survive contact with the real world. 🚗🤖

Breaking news: Cosmos 3 is here. They are attempting to do something completely new 🤯 Why is Physical AI much harder than building a chatbot? Understanding the world is not enough, robots need to predict it and act inside it. That's the idea behind NVIDIA Cosmos 3: → Reasoning model understands what's happening from video, images, text, and actions. → World model generates future states of the environment. → Action model generates the actions needed to achieve a goal. Previous systems often stitched these capabilities together using separate models. Cosmos 3 combines them into a single architecture with two components: ▪️ Reasoner Tower Analyzes observations and builds an understanding of objects, motion, interactions, and physical context. ▪️ Generator Tower Uses that understanding to generate future videos and action sequences that obey physical constraints. So Cosmos 3 moves from: Perception → Model A Prediction → Model B Actions → Model C to: Perception + Prediction + Actions → One unified system. The goal is to make robots, autonomous vehicles, and smart environments better at answering three questions: 1. What is happening? 2. What will happen next? 3. What should I do? That's a big shift from today's AI, which mostly focuses on generating text. And check the benchmarks! Physical AI needs to generate decisions that survive contact with the real world. 🚗🤖

Turing Post

13,755 次观看 • 2 个月前

WHAM defines the new state of the art in 3D human pose estimation from video. By a large margin. It’s fast, accurate, and it computes human pose in world coordinates. It’s also the first video-based method to be more accurate than single-image methods. 1/8

WHAM defines the new state of the art in 3D human pose estimation from video. By a large margin. It’s fast, accurate, and it computes human pose in world coordinates. It’s also the first video-based method to be more accurate than single-image methods. 1/8

Michael Black

118,463 次观看 • 2 年前

An interactive world model developed by NVIDIA in collaboration with academic partners. - DreamDojo turns egocentric human video data into physical intelligence. - Human data is more scalable than robotics data but lacks action labels. - To solve this, a dedicated action model extracts latent actions by identifying physics and motion deltas between frames. Training - A massive 44k hours of video data are used for pre-training. - Post-training on small-scale robot datasets maps human physics to specific robot embodiments. - An additional distillation stage converts the model into an autoregressive, few-step diffusion model, enabling real-time, action-controllable simulation. Primary Use Cases - Live Teleoperation: Controlling a robot inside a world simulation in real-time. - Model-based Planning: Previewing and curating the best actions for improved success. - Policy Evaluation: Testing robot policies in realistic, out-of-distribution scenarios. Everything that's open-sourced: weights, code, post-training dataset, eval set, and details to reproduce.

An interactive world model developed by NVIDIA in collaboration with academic partners. - DreamDojo turns egocentric human video data into physical intelligence. - Human data is more scalable than robotics data but lacks action labels. - To solve this, a dedicated action model extracts latent actions by identifying physics and motion deltas between frames. Training - A massive 44k hours of video data are used for pre-training. - Post-training on small-scale robot datasets maps human physics to specific robot embodiments. - An additional distillation stage converts the model into an autoregressive, few-step diffusion model, enabling real-time, action-controllable simulation. Primary Use Cases - Live Teleoperation: Controlling a robot inside a world simulation in real-time. - Model-based Planning: Previewing and curating the best actions for improved success. - Policy Evaluation: Testing robot policies in realistic, out-of-distribution scenarios. Everything that's open-sourced: weights, code, post-training dataset, eval set, and details to reproduce.

The Humanoid Hub

11,575 次观看 • 5 个月前

We open-source Action Images — a new representation that translates 7-DoF robot actions into interpretable images. Video models are emerging as powerful robotic foundation models, but a key challenge remains: how can we seamlessly integrate robot policies into video models? Instead of representing actions as low-dimensional control tokens, Action Images provide a pixel-grounded action representation, reframing policy learning as a visual tracking problem! By unifying observations and actions in the same video space, Action Images enable a unified robotics world model that supports video-action joint generation, action-conditioned video generation, and action labeling! Code: Paper:

We open-source Action Images — a new representation that translates 7-DoF robot actions into interpretable images. Video models are emerging as powerful robotic foundation models, but a key challenge remains: how can we seamlessly integrate robot policies into video models? Instead of representing actions as low-dimensional control tokens, Action Images provide a pixel-grounded action representation, reframing policy learning as a visual tracking problem! By unifying observations and actions in the same video space, Action Images enable a unified robotics world model that supports video-action joint generation, action-conditioned video generation, and action labeling! Code: Paper:

Chuang Gan

19,585 次观看 • 1 个月前

Matt Reeves says that he tried to make THE BATMAN like it would happen in real life. “I wanted this to be very much like it would happen in the real world. I wanted it to feel like it was grounded.” “like Chinatown in a certain sense. A description of a corrupt place and to get at the seeds of how far that corruption went… I wanted to have a story that feels very much of our world.”

Matt Reeves says that he tried to make THE BATMAN like it would happen in real life. “I wanted this to be very much like it would happen in the real world. I wanted it to feel like it was grounded.” “like Chinatown in a certain sense. A description of a corrupt place and to get at the seeds of how far that corruption went… I wanted to have a story that feels very much of our world.”

Emir Han

16,327 次观看 • 29 天前

Step into the map with the Street View grounding feature in Project Genie from Google DeepMind and Google Labs. Announced at I/O, this research prototype uses locations from Google Maps Street View as a foundation, letting you generate and explore interactive, 360-degree virtual environments from just a text prompt or real-world starting place. Cool, right? But… How does it actually work? 🤔 As an experimental tool, Project Genie tackles the "blank space" problem (showing both what’s in front of the camera and behind it) by utilizing Street View data to realistically generate a 360-degree view of the location you selected as the starting point for your world to generate from. Worlds generated by Genie are far more dynamic and rich because they’re created frame-by-frame based on the world description and user actions. By predicting each subsequent frame, Genie is able to simulate what it looks like to swim across an ocean, or hike to the top of a peak, marking a massive shift in interactive media and simulation pipelines. What real-world place would you want to step into and explore?

Step into the map with the Street View grounding feature in Project Genie from Google DeepMind and Google Labs. Announced at I/O, this research prototype uses locations from Google Maps Street View as a foundation, letting you generate and explore interactive, 360-degree virtual environments from just a text prompt or real-world starting place. Cool, right? But… How does it actually work? 🤔 As an experimental tool, Project Genie tackles the "blank space" problem (showing both what’s in front of the camera and behind it) by utilizing Street View data to realistically generate a 360-degree view of the location you selected as the starting point for your world to generate from. Worlds generated by Genie are far more dynamic and rich because they’re created frame-by-frame based on the world description and user actions. By predicting each subsequent frame, Genie is able to simulate what it looks like to swim across an ocean, or hike to the top of a peak, marking a massive shift in interactive media and simulation pipelines. What real-world place would you want to step into and explore?

Google AI

2,323,015 次观看 • 22 天前

I really want to show off the cutting edge tech the team is building and how we're turning AI image generation into real living environments using gaussian splats and at the :19 mark of this video you can see how we're blending splats together to create a real world. What this looks like from a gaming perspective is this all happens behind the players view to simulate a real world environment. As you move through the space, distance becomes a variable to consider on how images update and are experienced. We are creating a new reality in real time.

I really want to show off the cutting edge tech the team is building and how we're turning AI image generation into real living environments using gaussian splats and at the :19 mark of this video you can see how we're blending splats together to create a real world. What this looks like from a gaming perspective is this all happens behind the players view to simulate a real world environment. As you move through the space, distance becomes a variable to consider on how images update and are experienced. We are creating a new reality in real time.

𝔹𝕃𝕍ℂ𝕂𝕃!𝔾ℍ𝕋

25,084 次观看 • 1 个月前

One is ideal. The other is real. The ideal tells you how life should be. The real shows you how life is. The ideal is polished, perfect, inspiring. The real is messy, flawed, human. In this rare video, you might find yourself agreeing with both. And that’s the beauty of it, the ideal gives direction, but the real gives truth. If I had to choose, I would choose to be real. Because ideals can impress the world. But being real connects you to it. Ideals look good in theory. Reality builds character in practice. And in the long run, authenticity outlives perfection.

One is ideal. The other is real. The ideal tells you how life should be. The real shows you how life is. The ideal is polished, perfect, inspiring. The real is messy, flawed, human. In this rare video, you might find yourself agreeing with both. And that’s the beauty of it, the ideal gives direction, but the real gives truth. If I had to choose, I would choose to be real. Because ideals can impress the world. But being real connects you to it. Ideals look good in theory. Reality builds character in practice. And in the long run, authenticity outlives perfection.

Jitin Yadav, IAS

33,116 次观看 • 5 个月前

Greetings from the record holder for the longest space held on CTNG🔥 I dropped a short video to say thank you to everyone who showed up and held it down during my longest Space attempt 🤍🎙️ Your time, energy and support did not go unnoticed. It meant more than you know. Please like, repost and help this reach everyone who tuned in, so they know what went down and how real the support was 🔥we go again as soon as I get a response from Guinness World Records

Greetings from the record holder for the longest space held on CTNG🔥 I dropped a short video to say thank you to everyone who showed up and held it down during my longest Space attempt 🤍🎙️ Your time, energy and support did not go unnoticed. It meant more than you know. Please like, repost and help this reach everyone who tuned in, so they know what went down and how real the support was 🔥we go again as soon as I get a response from Guinness World Records

melissa

48,876 次观看 • 6 个月前

DisCo: Disentangled Control for Referring Human Dance Generation in Real World paper page: Generative AI has made significant strides in computer vision, particularly in image/video synthesis conditioned on text descriptions. Despite the advancements, it remains challenging especially in the generation of human-centric content such as dance synthesis. Existing dance synthesis methods struggle with the gap between synthesized content and real-world dance scenarios. In this paper, we define a new problem setting: Referring Human Dance Generation, which focuses on real-world dance scenarios with three important properties: (i) Faithfulness: the synthesis should retain the appearance of both human subject foreground and background from the reference image, and precisely follow the target pose; (ii) Generalizability: the model should generalize to unseen human subjects, backgrounds, and poses; (iii) Compositionality: it should allow for composition of seen/unseen subjects, backgrounds, and poses from different sources. To address these challenges, we introduce a novel approach, DISCO, which includes a novel model architecture with disentangled control to improve the faithfulness and compositionality of dance synthesis, and an effective human attribute pre-training for better generalizability to unseen humans. Extensive qualitative and quantitative results demonstrate that DISCO can generate high-quality human dance images and videos with diverse appearances and flexible motions.

DisCo: Disentangled Control for Referring Human Dance Generation in Real World paper page: Generative AI has made significant strides in computer vision, particularly in image/video synthesis conditioned on text descriptions. Despite the advancements, it remains challenging especially in the generation of human-centric content such as dance synthesis. Existing dance synthesis methods struggle with the gap between synthesized content and real-world dance scenarios. In this paper, we define a new problem setting: Referring Human Dance Generation, which focuses on real-world dance scenarios with three important properties: (i) Faithfulness: the synthesis should retain the appearance of both human subject foreground and background from the reference image, and precisely follow the target pose; (ii) Generalizability: the model should generalize to unseen human subjects, backgrounds, and poses; (iii) Compositionality: it should allow for composition of seen/unseen subjects, backgrounds, and poses from different sources. To address these challenges, we introduce a novel approach, DISCO, which includes a novel model architecture with disentangled control to improve the faithfulness and compositionality of dance synthesis, and an effective human attribute pre-training for better generalizability to unseen humans. Extensive qualitative and quantitative results demonstrate that DISCO can generate high-quality human dance images and videos with diverse appearances and flexible motions.

AK

161,478 次观看 • 3 年前

Vice President Kamala Harris has an opportunity to demonstrate our party’s courage and commitment to ending the genocide not just in condemning it with words, but in action by conditioning aid. A better world is possible, but we must have the courage to fight for it.

Vice President Kamala Harris has an opportunity to demonstrate our party’s courage and commitment to ending the genocide not just in condemning it with words, but in action by conditioning aid. A better world is possible, but we must have the courage to fight for it.

Ilhan Omar

298,838 次观看 • 1 年前

VLA-JEPA just dropped in LeRobot 🤖 What makes this model special is that it does not just learn what action to take from a given observation, it also leverages a JEPA world model to learn action-relevant dynamics. During training, the VLA leverages V-JEPA2 by conditioning its predictor. This clever trick adds a world modeling objective to the training, which also allows pretraining on human videos. At inference, the world model is dropped entirely, keeping only a standard VLA architecture: Qwen backbone and action head. The demo here was only fine-tuned on 13 examples, showing great pretraining capability and running in real time on NVIDIA Robotics DGX Spark! VLA-JEPA is the first world model to be ported to LeRobot, and I feel like it won't be the last 🚀 Thomas Wolf clem 🤗

VLA-JEPA just dropped in LeRobot 🤖 What makes this model special is that it does not just learn what action to take from a given observation, it also leverages a JEPA world model to learn action-relevant dynamics. During training, the VLA leverages V-JEPA2 by conditioning its predictor. This clever trick adds a world modeling objective to the training, which also allows pretraining on human videos. At inference, the world model is dropped entirely, keeping only a standard VLA architecture: Qwen backbone and action head. The demo here was only fine-tuned on 13 examples, showing great pretraining capability and running in real time on NVIDIA Robotics DGX Spark! VLA-JEPA is the first world model to be ported to LeRobot, and I feel like it won't be the last 🚀 Thomas Wolf clem 🤗

LeRobot

319,295 次观看 • 1 个月前

Not the flashiest demos, but what’s under the hood represents a foundational shift for general-purpose robotics. World models are the next-gen foundation of Physical AI, not the VLM backbones found in typical VLAs. DreamZero is a 14B-parameter World Action Model (WAM) by NVIDIA that treats robotics as a joint video-and-action prediction task. Unlike traditional Vision-Language-Action (VLA) models that map images directly to motor commands, DreamZero leverages a pretrained video diffusion backbone to predict future world states and actions simultaneously. - achieves 2× better zero-shot generalization to unseen tasks and environments compared to state-of-the-art VLAs. - learns effectively from heterogeneous, non-repetitive data (500 hours), breaking the need for thousands of repeated demonstrations. - adapts to new robot embodiments with just 30 minutes of play data. - enables 7Hz closed-loop control via system optimizations and "DreamZero-Flash," making high-capacity diffusion models viable for real-time use.

Not the flashiest demos, but what’s under the hood represents a foundational shift for general-purpose robotics. World models are the next-gen foundation of Physical AI, not the VLM backbones found in typical VLAs. DreamZero is a 14B-parameter World Action Model (WAM) by NVIDIA that treats robotics as a joint video-and-action prediction task. Unlike traditional Vision-Language-Action (VLA) models that map images directly to motor commands, DreamZero leverages a pretrained video diffusion backbone to predict future world states and actions simultaneously. - achieves 2× better zero-shot generalization to unseen tasks and environments compared to state-of-the-art VLAs. - learns effectively from heterogeneous, non-repetitive data (500 hours), breaking the need for thousands of repeated demonstrations. - adapts to new robot embodiments with just 30 minutes of play data. - enables 7Hz closed-loop control via system optimizations and "DreamZero-Flash," making high-capacity diffusion models viable for real-time use.

The Humanoid Hub

35,204 次观看 • 5 个月前