My main research
interests are in image processing, multi-view video
analysis. Specifically, my work concerns on using motion trajectories
to process video analysis. Click on a topic below for more
This work presents a novel approach to extract reliable dense and
long-range motion trajectories of articulated human in a video
sequence. Compared with existing approaches that emphasize temporal
consistency of each tracked point, we also consider the spatial
structure of tracked points on the articulated human. We treat points
as a set of vertex, and build a triangle mesh to join them in image
space. The problem of extracting long-range motion trajectories is
changed to the issue of consistency of mesh evolution over time.
Detection of moving object from monocular video captured by freely
moving camera is a challenging problem in computer vision. Unlike most
existing methods which impose significant geometric constraints or
background/foreground probabilistic modeling, this paper presents a
novel coarse-to-fine method based on motion compensation and adaptive
thresholding to address this detection problem.
By exploring the invariants of planar trajectories under
projective transformation, we proposed a novel approach to align
unsynchronized video sequences of the same dynamic scene that can be
subframe accurate and is applicable for different frame rate problem.
We proposed an
local-adaptation interpolation method for image super-resolution
reconstruction. Then we discussed how the quality of initial
interpolation image impact on iterative back projection (IBP) algorithm.