Office hours: CIT 521, TBD
Teaching:
CS1430, Introduction toComputer Vision, MWF 11:00-11:50, CIT 227
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 tracking;
the estimation of human body shape from images and video;
the representation and
detection of motion discontinuities;
the
estimation of optical flow.
I also do research on neural engineering for brain-machine interfaces and neural prostheses.
Recent Press:
Many uses seen for software that lays bare our 3-D selves, Boston Globe, Nov. 2008
Recent Vision Papers:
Human body shape:
Estimating human shape and pose from a single image (ICCV'09)
The naked truth: Estimating human shape under clothing
(ECCV'08)
Recovering human pose and shape in strong
lighting (ICCV'07)
Detailed human shape and pose from
images (CVPR'07)
Human tracking:
HumanEva: Synchronized video and motion capture dataset and baseline algorithm for evaluation of articulated human motion (IJCV'10)
Optical flow:
Learning optical flow (ECCV'08)
A database and evaluation methodology for optical flow,
(ICCV'07)
Spatial statistics of optical flow
(ICCV'05, Marr Prize honorable mention)
Markov random fields:
Steerable random fields
(ICCV'07)
Efficient belief propagation with learned
higher-order MRFs (ECCV 2006)
Fields of Experts (IJCV'09 version) [pdf from publisher]
Recent Papers in Neural Engineering
Neural control of computer cursor velocity by decoding motor cortical
spiking activity in humans with tetraplegia, (J. Neural Engineering '08)
A non-parametric Bayesian alternative to
spike sorting (J. Neuro Methods '08)
Consulting Activities
Scientific advisory board member of Videosurf [more]
Advisory board member of Willow Garage
|
