Matt Strong

@MattStr10050461

Stanford CS PhD | 3D vision for robotics and autonomy

peasant98.github.io

Joined December 2020

217 Following

193 Followers

Replies Reposts Media

Pinned
Matt Strong
@MattStr10050461
Oct 26
Can a robot handle coffee mugs and machines it has never seen, after having watched a human do the task once? Real2Sem2Real turns one phone video and a video scan of the objects into 2000+ robot demos across cousins, and matches or beats 150 teleop demos on 2 robots, across
8
Matt Strong
@MattStr10050461
Nov 4
@PGATour is this robot ready for the tour? Great work to @SwannAiden and team!
Aiden Swann 🤠 CORL2026
@SwannAiden
Nov 4
We taught a humanoid to play golf ⛳️ (audio on)
Matt Strong
@MattStr10050461
Oct 13
I think this work strikes a nice balance with generative and agentic. Lets generative do the initial heavy lifting, then refines with an agent. I think this will save a LOT of time for generation of sim-ready scenes, especially for robotics. I still find that estimating depth
Bowen Wen
@bowenwen_me
Oct 10
Casual photos in, simulation-ready 3D objects out! Introducing 🐊GATOR🐊, generative + agentic 3D object reconstruction. Here, humanoids interact with GATOR-reconstructed furniture [1/n] Website: research.nvidia.com/labs/lpr/gator/ Paper: arxiv.org/abs/2610.11215
Matt Strong
@MattStr10050461
Oct 5
Introducing VisTacAlign 👁️👆📐: co-training dexterous policies on human demos collected with a different tactile sensor than the robot’s. How we align vision and touch so cheap human data actually helps the robot! 🧵
2
Matt Strong
@MattStr10050461
Oct 1
Our lab has a new website! Check it out at arm.stanford.edu.
Home
From arm.stanford.edu

Matt Strong

Matt Strong’s posts