Yaser Sheikh

Timeline

1. 2001–2008

   PhD at UCF, postdoc at CMU

   He receives a BS from the Ghulam Ishaq Institute of Engineering Science and Technology in 2001 and the Hillman Fellowship for Excellence in Computer Science Research in 2004. In 2006, he earns his PhD at the University of Central Florida, advised by Mubarak Shah. In 2008, he completes a postdoctoral fellowship at Carnegie Mellon University under Takeo Kanade.

2. 2006–2019

   Faculty at Carnegie Mellon

   From 2006 to 2019, he is on the faculty of the Robotics Institute at Carnegie Mellon University, where he becomes an associate professor. His group develops OpenPose, a real-time 2D pose estimation approach, and the Panoptic Studio, a multisensor facility designed to capture social behavior. He wins the Honda Initiation Award in 2010.

3. 2015

   The Panoptic Studio paper

   He coauthors Panoptic Studio: A Massively Multiview System for Social Motion Capture for the International Conference on Computer Vision in 2015, with Hanbyul Joo, Takeo Kanade and others. The system takes 480 synchronized video streams of people engaged in social activities and produces the labeled 3D structure of anatomical landmarks on each person.

4. 2015–2025

   Leading Meta's Pittsburgh lab

   From 2015, he runs VR research for Facebook in Pittsburgh. As a Meta vice president until 2025, he founds the Meta Reality Lab in Pittsburgh and leads the invention and productization of Codec Avatars. Codec Avatars are digital representations of people designed for remotely sharing spaces with each other.

5. 2017

   Body and hand tracking at CVPR

   At CVPR 2017 in Honolulu, he and his colleagues present multiperson and hand-pose detection methods developed with the help of the Panoptic Studio, a two-story dome embedded with 500 video cameras. The group releases its code, and more than 20 commercial groups express interest in licensing the technology.

6. 2021

   Binaural speech at ICLR 2021

   ICLR 2021 places Neural Synthesis of Binaural Speech From Mono Audio, which he coauthors with Alexander Richard, Dejan Markovic and others, in its Outstanding Paper Session. The network turns a single-channel audio source into two-channel binaural sound, conditioned on the position and orientation of the listener relative to the source.

7. 2024

   URAvatar at SIGGRAPH Asia

   In 2024, he contributes to URAvatar: Universal Relightable Gaussian Codec Avatars, a technical paper at SIGGRAPH Asia, with a Reality Labs Research affiliation.

8. As of October 2026

   A stealth AI venture

   He is a consulting professor at Carnegie Mellon University and the founder of a new AI venture in stealth, focused on foundational advances in long-horizon foresight. His Meta years run from 2015 to 2025.