🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Check out our #ICRA2025 paper where we train a robotic puppy to dance expressively! Our method, Deep Fourier Mimic, a generalized version of DeepMimic, enables automatic parameterization of reference data. Project site: sony.github.io/DFM/ Video: youtube.com/watch?v=Do4HmC… 🧵

We introduce Deep Fourier Mimic, a generalized version of DeepMimic, which enables automatic parameterization of reference motions. This means you can learn diverse motions with a single policy conditioned on their meaningful spatial and temporal representations! #ICRA2025

Introducing Vega: 🤖 Dexmate's newest robot that makes complex manipulation tasks simple. ✨ A step closer to intelligence and automation. 🚀 🎥 Watch now: youtu.be/PecqfiJNwQI #Robotics #AI #Automation

🤖 Ever wondered what robots need to truly help humans around the house? 🏡 Introducing 𝗕𝗘𝗛𝗔𝗩𝗜𝗢𝗥 𝗥𝗼𝗯𝗼𝘁 𝗦𝘂𝗶𝘁𝗲 (𝗕𝗥𝗦)—a comprehensive framework for mastering mobile whole-body manipulation across diverse household tasks! 🧹🫧 From taking out the trash to

By treating humans and humanoids as interchangeable, we bridge costly teleoperation data with abundant, high-quality human data featuring minimal embodiment differences!

Teleoperation is so tedious. Can we find a better way to scale real-world data? Answer: Human videos. Inspiration: When we were doing teleoperation, we observed the motion performed by the human is almost the same as the robot, it is really just a 3D transformation away. Besides

Collect human data at scale using a VR app, then when you want to perform a task, you can predict human joints and retarget to humanoids. Really cool work

Human data ftw!

Today I was asked again on "Why humanoid?" The answer I gave is "It is the best embodiment to leverage human data" Introducing Humanoid Policy ~ Human Policy! A scalable way to collect and co-train humanoid policy with human data! human-as-robot.github.io Code and datasets are

[1/n] We are releasing M3 (#ICLR2025): a Gaussian Splatting method that builds LMM memories for arbitrary scenes. 🔥 [Efficient] 16 degrees in each Gaussian primitive for one LMM. 🔥 [Alignment] The rendered features are directly in the source LMM embedding space.

Check out our recent work on co-training humanoid policy with human data using a unified "human-centric state-action space": human-as-robot.github.io We also released a task-oriented egocentric dataset containing 27,000 human demos (3M frames) and 1,500 robot demos (0.7M frames)!

Test-Time Training (TTT) is now on Video! And not just a 5-second video. We can generate a full 1-min video! TTT module is an RNN module that provides an explicit and efficient memory mechanism. It models the hidden state of an RNN with a machine learning model, which is updated

Despite great advances in learning dexterity, hardware remains a major bottleneck. Most dexterous hands are either bulky, weak or expensive. I’m thrilled to present the RUKA Hand — a powerful, accessible research tool for dexterous manipulation that overcomes these limitations!

The current trend in human-demos-to-robot-policy papers reminds me of a time when many sim-to-real papers were focused on a particular theme: - You train a policy on the sim domain. - Then, during inference, you perform a domain transfer from real-to-sim to ensure consistency

Imagine robots learning new skills—without any robot data. Today, we're excited to release EgoZero: our first steps in training robot policies that operate in unseen environments, solely from data collected through humans wearing Aria smart glasses. 🧵👇