Joohwan Seo

University of California, Berkeley

Papers

8

Total Citations

61

H-Index

5

About

Joohwan Seo is a rising star at the intersection of geometric deep learning and robotic manipulation. His research centers on leveraging the mathematical structure of the Lie group SE(3) to create more intelligent, transferable, and physically-aware robot control systems. Seo’s major contributions lie in developing **SE(3)-equivariant** frameworks—both for learning and control—that allow robots to generalize manipulation skills across different poses and environments without retraining. His seminal work, *Diffusion-EDFs* (2024, 18 citations), introduces a bi-equivariant denoising generative model that learns stochastic human demonstrations for visual robotic manipulation, achieving state-of-the-art performance in contact-rich tasks. He has also pioneered **Geometric Impedance Control on SE(3)** (2023, 14 citations), providing a differential geometric foundation for compliant interaction with unknown environments. With over 60 cumulative citations and a comprehensive tutorial survey on SE(3)-equivariant robot learning (2025), Seo is establishing a unified geometric language for robot learning and control. His work promises to make robots not only more dexterous but also more principled in their understanding of physical space.

Research Focus

Key Achievements

5
H-Index
8
Papers
61
Total Citations
8
Avg Citations/Paper
🏆 Most Cited Paper
Diffusion-EDFs: Bi-Equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation
18 citations · 2024
📈 Most Prolific Year: 2023 (5 Papers)
🤝 Key Collaborators: 19
🏛 Institutions: University of California, Berkeley

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago