Office hours: CIT 521, Wed 3:00-4:00, Thurs 3:00-4:00.
Teaching:
CS26-4,
Topics in Computer Vision, W 3:00-5:20, Rm 506.
Research Interests
My Computer Vision Research focuses on estimating optical flow from sequences
of images. In particular I study
 the
statistics of natural images and image motion;
articulated
human motion estimation and full body tracking;
the representation and
detection of motion discontinuities;
the
estimation of optical flow and the recognition of motion
events;
high-dimensional robust learning and inference.
My research on Neural Engineering, Computational Neuroscience, and Brain-machine Interface
focuses on
statistical models of
neural coding;
probabilistic methods for
neural decoding;
developing neural
prostheses using implanted microelectrode arrays;
bionic
systems that couple brains and robots.
Call for papers:
IJCV Special Issue on Evaluation of Articulated Human
Motion and Pose Estimation (EHuM)Recent Vision Papers:
Recovering human pose and shape in strong
lighting (ICCV'07) <very cool>
Steerable random fields
(ICCV'07)
A database and evaluation methodology for optical flow,
(ICCV'07)
Detailed human shape and pose from
images (CVPR'07)
Efficient belief propagation with learned
higher-order MRFs (ECCV 2006)
Spatial statistics of optical flow
(ICCV'05, Marr Prize honorable mention)
Fields of Experts (CVPR'05)
Tracking loose-limbed people
(CVPR'04)
|
