Unsupervised categorization of objects is a fundamental problem in computer vision. While appearance-based methods have become popular recently, other important cues like function...
Global shape information is an effective top-down complement
to bottom-up figure-ground segmentation as well
as a useful constraint to avoid drift during adaptive tracking.
We p...
With the advance of high-speed network technologies, the availability and popularity of streaming media content over the Internet has grown rapidly in recent years. The delivery a...
In this paper we present a probabilistic framework for tracking objects based on local dynamic segmentation. We view the segn to be a Markov labeling process and abstract it as a ...
Most tracking algorithms implicitly apply a coarse segmentation of each target object using a simple mask such as a rectangle or an ellipse. Although convenient, such coarse segme...