معرفی
Yaser Ajmal Sheikh is an Associate Professor at the Robotics Institute of Carnegie Mellon University (currently on leave) and serves as Director of Facebook Reality Lab in Pittsburgh. His work focuses on "metric telepresence": remote interactions in AR/VR that are indistinguishable from reality. He has made significant contributions to machine perception and rendering of social behavior, spanning computer vision, computer graphics, and machine learning.
Dr. Sheikh's research encompasses analyzing dynamism in scenes from moving cameras, with particular focus on dynamic motion reconstruction, human behavior analysis, estimation of nonrigid motion, and modeling moving cameras in spacetime. His work has led to breakthroughs in social robotics, 3D vision and recognition, and multisensor data fusion. He founded and directs the Facebook Reality Lab in Pittsburgh, continuing his work on human-centered robotics and social robots.
His publication record demonstrates consistent innovation in computer vision and graphics, with emphasis on human motion capture, facial animation, and social interaction modeling. His most recent work focuses on neural approaches to facial animation, full-body avatars, and capturing the complex dynamics of human interaction. The research shows a clear trajectory from foundational work in motion capture to sophisticated neural approaches for realistic human representation in virtual environments.
- Popular Science's Best of What's New Award
- Honda Initiation Award (2010)
- Best paper awards at WACV (2012), SAP (2012), SCA (2010), ICCV THEMIS (2009)
- First place in the MSCOCO Keypoint Challenge (2016)
- Hillman Fellowship for Excellence in Computer Science Research (2004)
Dr. Sheikh has advised numerous doctoral and master's students who have gone on to prestigious positions in academia and industry. His research has been generously supported by government agencies including NSF and DARPA, as well as industrial partners such as Intel, Disney, Nissan, Honda, Toyota, and Samsung. He has served on senior committees at major conferences including SIGGRAPH, CVPR, ICRA, and ICCP, and was an Associate Editor of CVIU.
He leads the Panoptic Studio project, a massively multiview system for social motion capture that has produced the CMU Panoptic Studio Dataset. His work on OpenPose has become a standard tool in the field for real-time 2D hand, body, and face keypoint detection. His current research at Facebook Reality Lab continues to push the boundaries of what's possible in virtual and augmented reality through advanced computer vision techniques.
