Swaminathan Gurumurthy

Papers

Total Citations

H-Index

About

Swaminathan Gurumurthy is a leading researcher in legged robotics and imitation learning, best known for pioneering the transfer of complex, natural motions from casual video footage to robotic systems. His landmark work, "SLoMo: A General System for Legged Robot Motion Imitation From Casual Videos," introduces a groundbreaking three-stage framework that synthesizes physically plausible key-point trajectories from monocular, in-the-wild videos of humans and animals, enabling robots to replicate skilled movements without expensive motion-capture setups. This innovation has garnered 17 citations and is widely recognized for bridging the gap between unstructured visual data and robust robotic control. Gurumurthy’s contributions extend to advancing whole-body locomotion and dynamic motion synthesis, with his research directly impacting how robots learn agile behaviors from sparse, real-world observations. His work has been featured in top robotics venues, and he is celebrated for making robot motion imitation more accessible and scalable. By democratizing skill transfer from everyday videos, Gurumurthy is shaping a future where robots can fluidly adapt to diverse, unstructured environments, inspiring a new generation of researchers in embodied AI and autonomous systems.