Max Robotics — Global Service Robot Directory + US Certification

About

Jean-Baptiste Regli is a leading researcher at the intersection of computer vision and robotic manipulation, with a focus on how perception fundamentally constrains a robot’s ability to act. His most influential work, the S3K framework (Self-Supervised Semantic Keypoints), introduces a novel approach to visual representation learning that leverages multi-view consistency—enabling robots to perceive and manipulate objects without requiring large, labeled datasets. This work has garnered 14 citations and laid critical groundwork for self-supervised learning in robotics. Regli’s impact extends further with RoboCat, a self-improving generalist agent that can leverage heterogeneous robotic experience across different robots and tasks to quickly master novel skills and embodiments. This paper, with 9 citations, represents a significant step toward foundation models for robotics, analogous to advances in vision and language. By enabling multi-embodiment, multi-task generalization, Regli’s research addresses a core challenge in scaling robot learning. His contributions are shaping how robots perceive, adapt, and act in the real world, making him a key figure in the push toward more capable, general-purpose robotic systems.