Check out Silvia Zuffi’s latest work on 3D animals, which connects language with shape. arxiv.org/pdf/2404.03042…

🔔Can VLMs spatially refer objects in Ego? Can VLMs understand interactions? Which hand is holding an object and what's the object in the left hand? We show current VLMs struggle in interactions and release new data & models for HOI-Ref in Ego sid2697.github.io/hoi-ref/ On ArXiv🧵

🚀Next step in 3D scene understanding beyond objects and parts, we now explore interactive elements and functionalities👇 Check out our #CVPR2025 challenge and win2️⃣RTX4090🏅 (sponsored by Matterport🙏) 🛠️Details: opensun3d.github.io/cvpr24-challen… 📄Paper: scenefun3d.github.io #CVPR2024

Have CG art tools gotten to the point of modeling all the muscles in a human realistically, so a lot of the skinning and animation would be parametric variations on “baseline human” rather than free form constructions? I have occasionally advocated for that — some would complain

John Carmack I used to think that controlling movement through muscles was too complex but this ICLR paper changed my mind. But realism also requires modeling the subcutaneous and visceral adipose tissue (ie fat) and it's complex: more on that at #CVPR2024. github.com/martius-lab/de…

🙌🏻🙌🏻🙌🏻

Very excited to release OpenVLA today, a 7B parameter open-source vision-language-action model (VLA). 🦾 SoTA generalist policy (better than Octo & RT-2-X) ⚡️ Easy to run & fine-tune on 1 GPU with quantization and LoRA 💻 Open-source PyTorch codebase 🤗 Models on HuggingFace 1/

Wondering what people look like below the skin and fat? 💪🩻💀We introduce HIT: Human Implicit Tissues t.ly/PNJcF at #CVPR 🧵(1/7) 🚨I'm graduating soon and looking for opportunities, so let’s meet at #CVPR2024 📅 Wednesday, June 19 🕥 10:30 - 12:00 📍 Poster 4978

Come and check out our workshop MANGO @ summit 430. Happening now~ New Trends in Multimodal Human Action Perception, Understanding and Generation 🥭🥭🥭 mango-workshop.github.io/2024.html

Code of PuzzleAvatar (#SIGGRAPHAsia2024) gets released. 3D Human reconstruction from unconstrained photo collections (your album), in ANY poses, from ANY views, with ANY cropping or occlusion. NO estimation of body pose or camera is required. github.com/YuliangXiu/Puz…

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! homangab.github.io/gen2act/ 1/n

Excited to share my final PhD project😀 We show how simple, yet elegant changes enable diffusion transformers to learn SOTA robotic policies on real robots. Our method improves performance by 20% across a wide range of highly dexterous tasks - like cutting sushi! 1/n

Time to democratize humanoid robots! Introducing ToddlerBot, a low-cost ($6K), open-source humanoid for robotics and AI research. Watch two ToddlerBots seamlessly chain their loco-manipulation skills to collaborate in tidying up after a toy session. toddlerbot.github.io

We will be hosting the workshop on "Agents in Interactions, from Humans to Robots" at CVPR2025 #CVPR2025 , welcome to join us by submitting a paper or stopping by our talks/posters! For more info please check out: agents-in-interactions.github.io

Super impressive! A perfect example of how RL can accelerate robotics research, especially when with good hardware & infra: Previous Atlas demos (using TO + MPC) take years, now new Atlas RL demos only take months! Let me try to reverse-engineer the pipeline: 1. Some kind of

We're organizing a workshop on human motion modeling and learning robotic control by observing humans! Come join us #CVPR2025 in Nashville! We have an exciting group of speakers and are also inviting paper submissions with a May 7 deadline 1/n

Reminder that the deadline to submit your cool work to the Workshop on 4D vision @ CVPR 2025 is in about a week!!!

📢We are organizing the 1st workshop on "Agents in Interactions: from Humans👤to Robots🤖" at #CVPR2025 #CVPR2025! Welcome to join us by submitting your paper or hearing insightful talks from our excellent speakers!! Details: agents-in-interactions.github.io

Introducing St4RTrack!🖖 Simultaneous 4D Reconstruction and Tracking in the world coordinate feed-forwardly, just by changing the meaning of two pointmaps! st4rtrack.github.io

Human HOI data from Web helps functional grasp~