正在加载视频...

视频加载失败

返回首页

Excited to share that our paper 🌊🤺 “CFC: Simulating Character–Fluid Coupling using a Two-Level World Model” has been accepted to #SIGGRAPHASIA2025! In this work, we build a two-level world model (neural physics) for rigid-body–fluid interaction and use it to train physics-based character controllers efficiently. We study: (1) learning to...

frankzydou's profile picture

Zhiyang (Frank) Dou

1,993 subscribers

20,200 次观看 • 6 个月前 •via X (Twitter)

#SIGGRAPHASIA2025
LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Fluid physics with floating particles in real-time in #Blender3d with #geometrynodes It has 3 simulation passes - forces from rigid bodies (1) are sent to fluid solver (2) that are added on top of simulated water ripples (3). I find it quite meditative.
1:58

Fluid physics with floating particles in real-time in #Blender3d with #geometrynodes It has 3 simulation passes - forces from rigid bodies (1) are sent to fluid solver (2) that are added on top of simulated water ripples (3). I find it quite meditative.

Mārtiņš Upītis 🇱🇻 🇺🇦

142,537 次观看 • 2 年前

Today we are announcing Genie 3, a general purpose world model by Google DeepMind that can generate dynamic, interactive environments with a single text prompt. World models are AI that understand facets of the world (like Veo's knowledge of intuitive physics or Genie's mastery of new environments), and serve as a key stepping stone on the path to AGI. Genie 3 is our first world model to allow interaction in real-time, while also improving consistency and realism compared to Genie 2. Learn more ➡️
0:33

Today we are announcing Genie 3, a general purpose world model by Google DeepMind that can generate dynamic, interactive environments with a single text prompt. World models are AI that understand facets of the world (like Veo's knowledge of intuitive physics or Genie's mastery of new environments), and serve as a key stepping stone on the path to AGI. Genie 3 is our first world model to allow interaction in real-time, while also improving consistency and realism compared to Genie 2. Learn more ➡️

Google AI

102,524 次观看 • 11 个月前

New video series: Physics Informed Machine Learning! Physics may be embedded into AI/ML in 5 stages: 1 choose what to Model 2 curate training Data 3 design an Architecture 4 craft a Loss Function, and 5 implement Optimization Algorithm to train the model
2:19

New video series: Physics Informed Machine Learning! Physics may be embedded into AI/ML in 5 stages: 1 choose what to Model 2 curate training Data 3 design an Architecture 4 craft a Loss Function, and 5 implement Optimization Algorithm to train the model

Steven Brunton

78,441 次观看 • 2 年前

Why not use the real-world as a game level? This is a real-time augmented reality prototype using a physics based character controller, all running on the web 👀
0:30

Why not use the real-world as a game level? This is a real-time augmented reality prototype using a physics based character controller, all running on the web 👀

Ian Curtis

533,309 次观看 • 2 年前

Basic SPH fluid simulation using the 'for-each-element' zone in #b3d. Simulation is based on Muller's paper TenMinutePhysics #geometrynodes #animation #3dartwork #simulation #physics #science #educational #engineering
0:25

Basic SPH fluid simulation using the 'for-each-element' zone in #b3d. Simulation is based on Muller's paper TenMinutePhysics #geometrynodes #animation #3dartwork #simulation #physics #science #educational #engineering

Sam M

10,333 次观看 • 1 年前

Introducing CycleQD: A population-based model merging via Quality Diversity CycleQD builds on our model merging research, advancing two fronts: evolving a swarm of specialized agents to complement one another, and laying the groundwork for life-long learning by enabling diverse, adaptable skill acquisition at the population-level.
1:24

Introducing CycleQD: A population-based model merging via Quality Diversity CycleQD builds on our model merging research, advancing two fronts: evolving a swarm of specialized agents to complement one another, and laying the groundwork for life-long learning by enabling diverse, adaptable skill acquisition at the population-level.

Sakana AI

121,567 次观看 • 1 年前

We RL'ed humanoid robots to Cristiano Ronaldo, LeBron James, and Kobe Byrant! These are neural nets running on real hardware at our GEAR lab. Most robot demos you see online speed videos up. We actually *slow them down* so you can enjoy the fluid motions. I'm excited to announce "ASAP", a "real2sim2real" model that masters extremely smooth and dynamic motions for humanoid whole body control. We pretrain the robot in simulation first, but there is a notorious "sim2real" gap: it's very difficult for hand-engineered physics equations to match real world dynamics. Our fix is simple: just deploy a pretrained policy on real hardware, collect data, and replay the motion in sim. The replay will obviously have many errors, but that gives a rich signal to compensate for the physics discrepancy. Use another neural net to learn the delta. Basically, we "patch up" a traditional physics engine, so that the robot can experience almost the real world at scale in GPUs. The future is hybrid simulation: combine the power of classical sim engines refined over decades and the uncanny ability of modern NNs to capture a messy world.
0:50

We RL'ed humanoid robots to Cristiano Ronaldo, LeBron James, and Kobe Byrant! These are neural nets running on real hardware at our GEAR lab. Most robot demos you see online speed videos up. We actually *slow them down* so you can enjoy the fluid motions. I'm excited to announce "ASAP", a "real2sim2real" model that masters extremely smooth and dynamic motions for humanoid whole body control. We pretrain the robot in simulation first, but there is a notorious "sim2real" gap: it's very difficult for hand-engineered physics equations to match real world dynamics. Our fix is simple: just deploy a pretrained policy on real hardware, collect data, and replay the motion in sim. The replay will obviously have many errors, but that gives a rich signal to compensate for the physics discrepancy. Use another neural net to learn the delta. Basically, we "patch up" a traditional physics engine, so that the robot can experience almost the real world at scale in GPUs. The future is hybrid simulation: combine the power of classical sim engines refined over decades and the uncanny ability of modern NNs to capture a messy world.

Jim Fan

566,520 次观看 • 1 年前

Using Computational Physics and Fluid Dynamics to Imitate Biology
1:00

Using Computational Physics and Fluid Dynamics to Imitate Biology

Mathelirium

24,291 次观看 • 4 个月前

Released a new physics simulation - "Floaty"! Two-way coupling of fluid & soft body is simulated using Position Based Dynamics (PBD). I love how buoyancy is simulated! Optimized using Web Worker. Demo: Repo:
0:28

Released a new physics simulation - "Floaty"! Two-way coupling of fluid & soft body is simulated using Position Based Dynamics (PBD). I love how buoyancy is simulated! Optimized using Web Worker. Demo: Repo:

matsuoka-601

15,918 次观看 • 10 个月前

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!
53:45

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 • 22 天前

Fluid football from Blyth Spartans to level up the tie 😮‍💨 #EmiratesFACup
0:36

Fluid football from Blyth Spartans to level up the tie 😮‍💨 #EmiratesFACup

Emirates FA Cup

39,157 次观看 • 1 年前

🚨 New Paper Hybrid Neural World Models. We build neural world models for discontinuous environments and our method discovers when they're making error, with no ground truth. Our neural models run physics fast, leaping many simulator steps in a single pass. How we find points where not to trust it: ask the model for the same future two ways, and the disagreement between the two answers marks where it's failing without needing labels. The same recipe holds on three systems with nothing in common: a chemical wave, compressible gas with shocks, rigid bodies colliding in 3D. 👇
0:46

🚨 New Paper Hybrid Neural World Models. We build neural world models for discontinuous environments and our method discovers when they're making error, with no ground truth. Our neural models run physics fast, leaping many simulator steps in a single pass. How we find points where not to trust it: ask the model for the same future two ways, and the disagreement between the two answers marks where it's failing without needing labels. The same recipe holds on three systems with nothing in common: a chemical wave, compressible gas with shocks, rigid bodies colliding in 3D. 👇

Lossfunk

14,828 次观看 • 1 个月前

I've had early access to Kling AI brand new 2.0 video model and done extensive testing and trust me when I say that this model upgrade is AMAZING!!! 🔥 The amount of incredible dynamic action that you can generate now is next level. If you want to make things go fast, the new model completely excels at this, movement looks very natural and fluid. This is a major release for Kling! Here is a collection of some clips I've generated with the 2.0 model.
1:09

I've had early access to Kling AI brand new 2.0 video model and done extensive testing and trust me when I say that this model upgrade is AMAZING!!! 🔥 The amount of incredible dynamic action that you can generate now is next level. If you want to make things go fast, the new model completely excels at this, movement looks very natural and fluid. This is a major release for Kling! Here is a collection of some clips I've generated with the 2.0 model.

Travis Davids

87,789 次观看 • 1 年前

Im LITERALLY just trying to show the IMPRESSIVE and FLUID physics of the sand and water at the beach. It is incredibly ARTISTIC that games have reached this level. Western Devs would rather spend time making another HIDEOUS protagonist
0:24

Im LITERALLY just trying to show the IMPRESSIVE and FLUID physics of the sand and water at the beach. It is incredibly ARTISTIC that games have reached this level. Western Devs would rather spend time making another HIDEOUS protagonist

Windpress

68,028 次观看 • 2 个月前

Super excited to share the last paper of my PhD: "Hallucination in World Models is Predictable and Preventable"✨ We train a 350M-param generative world model on a large dataset w/ 210 tasks and show that we can predict *when* hallucination happens and use that to fix it! 🧵1/n
0:40

Super excited to share the last paper of my PhD: "Hallucination in World Models is Predictable and Preventable"✨ We train a 350M-param generative world model on a large dataset w/ 210 tasks and show that we can predict *when* hallucination happens and use that to fix it! 🧵1/n

Nicklas Hansen

50,109 次观看 • 8 天前

V7 of the EDCO fluid sim is complete. I declare this model a stunning success - I appear to have achieved parity in the inundation between the two projections. Recall that Blender's liquid mesh is an additional visualization based on the underlying particle system.
0:25

V7 of the EDCO fluid sim is complete. I declare this model a stunning success - I appear to have achieved parity in the inundation between the two projections. Recall that Blender's liquid mesh is an additional visualization based on the underlying particle system.

Junho

81,980 次观看 • 8 个月前

Fluid Markets on Polygon | POL are live on Superlend! Explore, deposit and earn yield directly on Fluid 🌊 via Superlend's interface.
0:33

Fluid Markets on Polygon | POL are live on Superlend! Explore, deposit and earn yield directly on Fluid 🌊 via Superlend's interface.

Superlend

37,316 次观看 • 1 年前

A droplet “walking” on a vibrating fluid can mimic quantum physics, forming diffraction patterns just like electrons in the Kapitza-Dirac effect. A rare, visible model of wave-particle duality, letting us watch quantum-like behavior in the macroscopic world.
2:01

A droplet “walking” on a vibrating fluid can mimic quantum physics, forming diffraction patterns just like electrons in the Kapitza-Dirac effect. A rare, visible model of wave-particle duality, letting us watch quantum-like behavior in the macroscopic world.

Bluntly Put Philosopher (BPP)

16,439 次观看 • 7 个月前

BRUH. The effects, the fluid fabric and character acting. This cut is full on insane.
1:08

BRUH. The effects, the fluid fabric and character acting. This cut is full on insane.

A̸c̸k̸e̸r̸m̸a̸n̸♨️

2,256,950 次观看 • 2 年前