Loading video...

Video Failed to Load

Go Home

What representation enables open-world robot manipulation from generated videos? Introducing Dream2Flow, our recent work that bridges video generation and robot control with 3D object flow. Stanford University #ICRA2026 1/N

wenlong_huang's profile picture

Wenlong Huang

5,667 subscribers

105,597 views • 3 months ago •via X (Twitter)

Science & Technology#ICRA2026
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Dream2Flow Bridging Video Generation and Open-World Manipulation with 3D Object Flow
2:59

Dream2Flow Bridging Video Generation and Open-World Manipulation with 3D Object Flow

AK

29,980 views • 6 months ago

Gen2Act: Casting language-conditioned manipulation as *human video generation* followed by *closed-loop policy execution conditioned on the generated video* enables solving diverse real-world tasks unseen in the robot dataset! 1/n
0:28

Gen2Act: Casting language-conditioned manipulation as *human video generation* followed by *closed-loop policy execution conditioned on the generated video* enables solving diverse real-world tasks unseen in the robot dataset! 1/n

Homanga Bharadhwaj

71,143 views • 1 year ago

Can we collect robot data without any robots? Introducing Universal Manipulation Interface (UMI) An open-source $400 system from Stanford University designed to democratize robot data collection 0 teleop -> autonomously wash dishes (precise), toss (dynamic), and fold clothes (bimanual)
1:16

Can we collect robot data without any robots? Introducing Universal Manipulation Interface (UMI) An open-source $400 system from Stanford University designed to democratize robot data collection 0 teleop -> autonomously wash dishes (precise), toss (dynamic), and fold clothes (bimanual)

Cheng Chi

438,741 views • 2 years ago

What if we can simulate an *interactive 3D world*, from a single image, in the wild, in real time? Introducing PointWorld-1B: a large pre-trained 3D world model that predicts env dynamics given RGB-D capture and robot actions. 🌐 from Stanford University NVIDIA
2:08

What if we can simulate an *interactive 3D world*, from a single image, in the wild, in real time? Introducing PointWorld-1B: a large pre-trained 3D world model that predicts env dynamics given RGB-D capture and robot actions. 🌐 from Stanford University NVIDIA

Wenlong Huang

273,346 views • 5 months ago

Introducing Yell At Your Robot (YAY Robot!) 🗣️- a fun collaboration b/w Stanford University and UC Berkeley 🤖 We enable robots to improve on-the-fly from language corrections: robots rapidly adapt in real-time and continuously improve from human verbal feedback. YAY Robot enables long-horizon, dexterous manipulation tasks like preparing trail-mix, packing a ziploc bag, and cleaning dishes:
2:58

Introducing Yell At Your Robot (YAY Robot!) 🗣️- a fun collaboration b/w Stanford University and UC Berkeley 🤖 We enable robots to improve on-the-fly from language corrections: robots rapidly adapt in real-time and continuously improve from human verbal feedback. YAY Robot enables long-horizon, dexterous manipulation tasks like preparing trail-mix, packing a ziploc bag, and cleaning dishes:

Lucy Shi

122,827 views • 2 years ago

Humans grasp objects with a purpose! Web2Grasp enables such functional grasping for dexterous robot hands via hand-object reconstruction from web images - without *any* robot teleop data collection 1/n
0:49

Humans grasp objects with a purpose! Web2Grasp enables such functional grasping for dexterous robot hands via hand-object reconstruction from web images - without *any* robot teleop data collection 1/n

Homanga Bharadhwaj @ CVPR

28,408 views • 1 year ago

Introducing a way to convert synthesized robot videos to robot execution without using any action labels! We also release a codebase (with pretrained models) for text-to-video generation. Train your own models for robot control in only 1 day with 4 GPUs!
1:51

Introducing a way to convert synthesized robot videos to robot execution without using any action labels! We also release a codebase (with pretrained models) for text-to-video generation. Train your own models for robot control in only 1 day with 4 GPUs!

Yilun Du

20,949 views • 2 years ago

How to harness foundation models for *generalization in the wild* in robot manipulation? Introducing VoxPoser: use LLM+VLM to label affordances and constraints directly in 3D perceptual space for zero-shot robot manipulation in the real world! 🌐 🧵👇
0:50

How to harness foundation models for *generalization in the wild* in robot manipulation? Introducing VoxPoser: use LLM+VLM to label affordances and constraints directly in 3D perceptual space for zero-shot robot manipulation in the real world! 🌐 🧵👇

Wenlong Huang

293,876 views • 3 years ago

So I heard we need more data for robot learning :) Purely real world teleop is expensive and slow, making large scale data collection challenging. I’ve been excited about getting more data into robot learning, going beyond just real-world teleop data. To this end, we’ve been scaling up data generation with RL in realistic simulations generated on the fly from crowdsourced videos. Enables realistic data collection, much more cheaply than purely real world teleop. Importantly, data collection becomes even*cheaper* with more environments, allowing training with over 100x more data. Transfers to real robots for generalizable manipulation. A 🧵 (1/N)
0:41

So I heard we need more data for robot learning :) Purely real world teleop is expensive and slow, making large scale data collection challenging. I’ve been excited about getting more data into robot learning, going beyond just real-world teleop data. To this end, we’ve been scaling up data generation with RL in realistic simulations generated on the fly from crowdsourced videos. Enables realistic data collection, much more cheaply than purely real world teleop. Importantly, data collection becomes even*cheaper* with more environments, allowing training with over 100x more data. Transfers to real robots for generalizable manipulation. A 🧵 (1/N)

Abhishek Gupta

13,346 views • 1 year ago

🤖What if a robot could perform a new task just from a natural language command, with zero demonstrations? Our new work, NovaFlow, makes it possible! We use pre-trained video generative model to create a video of the task, then translate it into a plan for real-world robot execution. 1/6 #Robotics #AI #ZeroShot #Manipulation
0:27

🤖What if a robot could perform a new task just from a natural language command, with zero demonstrations? Our new work, NovaFlow, makes it possible! We use pre-trained video generative model to create a video of the task, then translate it into a plan for real-world robot execution. 1/6 #Robotics #AI #ZeroShot #Manipulation

Hongyu Li

105,526 views • 8 months ago

Chinese #robot is dancing on #Stanford University campus.
1:37

Chinese #robot is dancing on #Stanford University campus.

Chinese Embassy in US

22,420 views • 10 months ago

Is VideoGen starting to become good enough for robotic manipulation? 🤖 Check out our recent work, RIGVid — Robots Imitating Generated Videos — where we use AI-generated videos as intermediate representations and 6-DoF motion retargeting to guide robots in diverse manipulation tasks: pouring, wiping, mixing, and more. 🔗 Key takeaways: - VideoGen starts to become good enough for robotics - As the field progresses, we are expecting much better results in the coming years - Depending on whether video prediction models take actions or not (VideoGen vs Action-Conditioned Video Prediction), there are different ways to use them. - Controllability & steerability are still issues In the paper, we explore: – How do different video generation models compare for robotic imitation? – Can generated videos replace real videos for imitation? – What causes failure of imitation given high-quality videos? – How does imitating from video compare with other representations (e.g., keypoint constraints like ReKep)? 🎥 Watch the video for (1) AI-generated inputs, (2) robot executions, and (3) the 3D intermediate representation bridging the embodiment gap.
0:39

Is VideoGen starting to become good enough for robotic manipulation? 🤖 Check out our recent work, RIGVid — Robots Imitating Generated Videos — where we use AI-generated videos as intermediate representations and 6-DoF motion retargeting to guide robots in diverse manipulation tasks: pouring, wiping, mixing, and more. 🔗 Key takeaways: - VideoGen starts to become good enough for robotics - As the field progresses, we are expecting much better results in the coming years - Depending on whether video prediction models take actions or not (VideoGen vs Action-Conditioned Video Prediction), there are different ways to use them. - Controllability & steerability are still issues In the paper, we explore: – How do different video generation models compare for robotic imitation? – Can generated videos replace real videos for imitation? – What causes failure of imitation given high-quality videos? – How does imitating from video compare with other representations (e.g., keypoint constraints like ReKep)? 🎥 Watch the video for (1) AI-generated inputs, (2) robot executions, and (3) the 3D intermediate representation bridging the embodiment gap.

Yunzhu Li

16,534 views • 1 year ago

Introducing MilliVid, our new method for long-context video generation! MilliVid creates videos that are consistent over long time spans, without using retrieval heuristics or 3D maps! (1/n)
0:32

Introducing MilliVid, our new method for long-context video generation! MilliVid creates videos that are consistent over long time spans, without using retrieval heuristics or 3D maps! (1/n)

Vincent Sitzmann

55,055 views • 26 days ago

The future of robot butlers starts with mobile manipulation. We’re announcing the NeurIPS 2023 Open-Vocabulary Mobile Manipulation Challenge! - Full robot stack ✅ - Parallel sim and real evaluation ✅ - No robot required ✅👀
1:24

The future of robot butlers starts with mobile manipulation. We’re announcing the NeurIPS 2023 Open-Vocabulary Mobile Manipulation Challenge! - Full robot stack ✅ - Parallel sim and real evaluation ✅ - No robot required ✅👀

Chris Paxton

178,966 views • 3 years ago

State-of-the-art robot policies often need hundreds of hours of data. What if we needed none? Introducing TiPToP: a manipulation system that zero-shots open-world tasks from pixels and language using vision foundation models and GPU-parallelized Task and Motion Planning (TAMP).
3:43

State-of-the-art robot policies often need hundreds of hours of data. What if we needed none? Introducing TiPToP: a manipulation system that zero-shots open-world tasks from pixels and language using vision foundation models and GPU-parallelized Task and Motion Planning (TAMP).

Nishanth Kumar

77,488 views • 3 months ago

Introducing Neural Jacobian Fields, robot 3D kinematic models learned only from vision! They can model & control robots from just a single RGB camera, even those w/ intractable kinematics & no embedded sensors such as soft, 3D-printed pneumatic hands! 1/n
0:55

Introducing Neural Jacobian Fields, robot 3D kinematic models learned only from vision! They can model & control robots from just a single RGB camera, even those w/ intractable kinematics & no embedded sensors such as soft, 3D-printed pneumatic hands! 1/n

Vincent Sitzmann

54,204 views • 1 year ago

AGIBOT founder Zhihui Peng unveiled the Lingxi X2, a bipedal humanoid robot integrating motion, interaction, and task intelligence. With 28 degrees of freedom, a Diffusion-based motion engine, and a computing system that enables natural human-robot interaction. It can walk, run, dance, and ride a bicycle, while also excelling in precise object manipulation.
12:03

AGIBOT founder Zhihui Peng unveiled the Lingxi X2, a bipedal humanoid robot integrating motion, interaction, and task intelligence. With 28 degrees of freedom, a Diffusion-based motion engine, and a computing system that enables natural human-robot interaction. It can walk, run, dance, and ride a bicycle, while also excelling in precise object manipulation.

The Humanoid Hub

124,533 views • 1 year ago

Introducing MicroFactory A robot that automates repetitive manual work — starting with electronics assembly.
1:24

Introducing MicroFactory A robot that automates repetitive manual work — starting with electronics assembly.

Igor Kulakov (MicroFactory)

719,531 views • 1 year ago

How to use simulation data for real-world robot manipulation? We present sim-and-real co-training, a simple recipe for manipulation. We demonstrate that sim data can significantly enhance real-world performance, even with notable differences between the sim and the real. (1/n)
1:15

How to use simulation data for real-world robot manipulation? We present sim-and-real co-training, a simple recipe for manipulation. We demonstrate that sim data can significantly enhance real-world performance, even with notable differences between the sim and the real. (1/n)

Zhenyu Jiang

44,263 views • 1 year ago

I'm excited to share our new work Diffusion as Shader (DaS), a versatile controllable video generation method for various tasks: object manipulation, camera control, mesh-to-video, and motion transfer. Project page: Github:
0:46

I'm excited to share our new work Diffusion as Shader (DaS), a versatile controllable video generation method for various tasks: object manipulation, camera control, mesh-to-video, and motion transfer. Project page: Github:

Yuan Liu

33,363 views • 1 year ago