Anirudh Pai

EECS M.S @ U.C Berkeley BAIR
Computer Vision & Dexterous Robotic Manipulation
Prev. at Meta, Two Sigma

[Roei Herzig @roeiherzig]

Direct map comparison between SigLIP, DINOv2, and CAIP...

Action-centric Vs. Semantic/low-level representations

@CVPR

[Roei Herzig @roeiherzig]

https://t.co/UOBRQXqyoX

[Roei Herzig @roeiherzig]

Children learn to manipulate the world by playing with toys — can robots do the same? 🧸🤖

We show that robots trained on 250 "toys" made of 4 shape primitives (🔵,🔶,🧱,💍) can generalize grasping to real objects.

@JitendraMalikCV @trevordarrell Shankar Sastry @berkeley_ai😊