
Yaser Ajmal Sheikh
Assistant Research Professor
Robotics Institute
School of Computer Science
Carnegie Mellon University
5000 Forbes Avenue
Pittsburgh, PA 15213
Email: myfirstname@cs.cmu.edu
Phone: (412) 268-1138
Office: NSH 4531
Doctoral Students
Ankur Datta (co-advised with Takeo Kanade)
Ijaz Akhter (co-advised with Sohaib Khan)
Eakta Jain (co-advised with Jessica Hodgins)
Hyun Soo Park
Natasha Kholgade
|

[ Complete List of Publications ] [ DBLP List ] [ Google Scholar ] |
Recent Publications |
 |
NEW! Modeling the Product Manifold of Posture and Motion
Ankur Datta, Yaser Sheikh, and Takeo Kanade
Workshop on Tracking Humans for the Evaluation of their Motion in Image Sequences (THEMIS), 2009 (held in conjunction with ICCV).
Best Paper Award Winner! |
 |
NEW! Dynamic Seethroughs: Synthesizing Hidden Views of Moving Objects
Peter Barnum, Yaser Sheikh, Ankur Datta, and Takeo Kanade
International Symposium on Multimedia and Augmented Reality (ISMAR), 2009.
Project Page (includes video): [ url ]
Press Coverage: Augmented reality system lets you see through walls, New Scientist Magazine, October 2009. |
 |
Leveraging the Talent of Hand Animators to Create Three-Dimensional Animation
Eakta Jain, Yaser Sheikh, and Jessica Hodgins
Symposium on Computer Animation (SCA), 2009. |
 |
Background Subtraction for Freely Moving Cameras
Yaser Sheikh, Omar Javed, and Takeo Kanade
IEEE International Conference on Computer Vision (ICCV), 2009. |
 |
Matching Trajectories of Anatomical Landmarks under Viewpoint Anthropometric, and Temporal Transforms
Alexei Gritai, Yaser Sheikh, Cen Rao, and Mubarak Shah
International Journal of Computer Vision (IJCV), 2009. |
 |
In Defense of Orthonormality Constraints for Nonrigid Structure from Motion (oral)
Ijaz Akhter, Yaser Sheikh, and Sohaib Khan
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2009. |
 |
Nonrigid Structure from Motion in Trajectory Space (oral)
Ijaz Akhter, Yaser Sheikh, Sohaib Khan, and Takeo Kanade
Neural Information Processing Systems (NIPS), 2008.
Matlab Code: [ url ] (includes dataset)
Video: [ url ] |
 |
Trajectory Association Across Multiple Airborne Cameras
Yaser Sheikh and Mubarak Shah
IEEE Transactions on Pattern Analysis and Machine Intelligence
(TPAMI), 2008.
IEEE International Conference on Computer Vision (ICCV) 2005.
|
 |
Linear Motion Estimation for Articulated Planes
Ankur Datta, Yaser Sheikh, and Takeo Kanade,
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2008.
Matlab Code: [ zip ] Video: [ mov ] |
  |
Mode-seeking by Medoidshifts (oral)
Yaser Sheikh, Erum Khan, and Takeo
Kanade
IEEE International Conference on Computer Vision (ICCV), 2007.
Matlab Code: [ zip ]
(Isomap code from http://isomap.stanford.edu/) |
 |
On the Spacetime Geometry of Galilean Cameras
Yaser Sheikh, Alexei Gritai, and Mubarak Shah,
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2007. |
 |
Shape from Dynamic Textures for Planes
Yaser Sheikh, Niels Haering, and Mubarak Shah
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2006. |
 |
Background Modeling in Dynamic Scenes
Yaser Sheikh and Mubarak Shah
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2005.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2005.
Matlab Code: [ zip ]
Data: [ rar ] (includes "ground truthed" data) |
|
Research Focus
My research focuses on understanding real world scenes from cameras. Specifically, the following three themes span most of my work:
-
Distributed Visual Understanding: The proliferation of cameras in everyday life, first with camcorders and now with cellphone cameras, is introducing many interesting computational paradigms. How can we best understand the world with these hundreds of millions of cameras that see everything? I'm leading the construction of the new Virtualizing Engine with Takeo Kanade to investigate this line of research.
-
Dynamic Scene Analysis: The interaction, in video, of motion induced by moving cameras and and the motion induced by dynamic objects, in particular, intrigues me. I'm working on developing methods to reconstruct dynamic scenes from moving cameras.
-
Human Activity Analysis: I am interested in inferring human intention from actions, exploring the tendency of people to behave similarly in similar situations. This research thread includes detection, tracking, and reconstructing people in images and video.
Sponsors
Press Coverage
Augmented reality system lets you see through walls
New Scientist Magazine, October 2009.
Other Links
|