Hanbyul Joo

Hanbyul Joo is an Assistant Professor in the Department of Computer Science and Engineering at Seoul National University (SNU). Prior to joining SNU, he was a Research Scientist at Facebook AI Research (FAIR) in Menlo Park. He completed his Ph.D. in the Robotics Institute at Carnegie Mellon University, working with Yaser Sheikh, and received his M.S. in Electrical Engineering and B.S. in Computer Science from KAIST, Korea.

Dr. Joo's educational journey began at KAIST, where he earned both his Bachelor's and Master's degrees. He then pursued his Ph.D. at Carnegie Mellon University's Robotics Institute, completing his dissertation titled "Sensing, Measuring, and Modeling Social Signals in Nonverbal Communication." His doctoral work focused on developing the Panoptic Studio, a unique sensing system with over 500 synchronized cameras for capturing social interactions.

Dr. Joo's research primarily focuses on endowing machines and robots with the ability to perceive and understand human behaviors in 3D. His goal is to build "social Artificial Intelligence" that can interact with humans using social signals (body languages). He pursues this direction using data-driven methods where data is collected by measuring the wide spectrum of social signals transmitted during interpersonal social interaction. His research spans computer vision, machine learning, computer graphics, and robotics, with particular emphasis on 3D human pose estimation, human-object interaction, and social signal processing.

His recent publications demonstrate a clear trend toward leveraging diffusion models for 3D reconstruction and generation tasks, with a focus on human-centric applications. His work bridges the gap between 2D image understanding and 3D scene reconstruction, often utilizing pre-trained models to overcome data limitations. The research consistently addresses fundamental challenges in understanding human behavior, interaction with objects, and social dynamics in 3D space.

Dr. Joo is a recipient of several prestigious awards including the Samsung Scholarship and the CVPR Best Student Paper Award in 2018. His paper "Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodies" received this honor at CVPR 2018. His research has been widely recognized in top computer vision and AI conferences, with multiple oral presentations at venues like CVPR, ICCV, and ECCV.

Dr. Joo actively mentors a large group of students, with approximately 15 current students working toward MS/PhD degrees under his supervision. His lab, the SNU VCLab, focuses on cutting-edge research in computer vision and graphics. He has secured significant research funding through his work, though specific grant details aren't provided on his website. Dr. Joo frequently serves as an area chair for major conferences including CVPR, ICCV, and NeurIPS, demonstrating his standing in the academic community.

Dr. Joo leads the SNU VCLab, which has developed several notable datasets and tools including SNU ParaHome, FrankMocap, and the CMU Panoptic Studio Dataset. His lab maintains strong industry connections, with students interning at leading companies like Meta. The lab's research focuses on building the infrastructure and algorithms needed for social AI, with an emphasis on practical applications that can be deployed in real-world settings.