Most of current machine vision systems suffer from a lack of flexibility to account for the high variability of unstructured environments. Here, as the state of the world evolves ...
Abstract. This paper presents a novel probabilistic approach to integrating multiple cues in visual tracking. We perform tracking in different cues by interacting processes. Each p...
A new method for object tracking in video sequences is presented. This method exploits the benefits of particle filters to tackle the multimodal distributions emerging from clutte...
Alexandros Makris, Dimitrios I. Kosmopoulos, Stavr...
Robust visual tracking has become an important topic in the field of computer vision. The integration of cues such as color, edge strength and motion has proved to be a promising ...
Chunhua Shen, Anton van den Hengel, Anthony R. Dic...
An attractive approach to improve tracking performance for visual surveillance is to use information from multiple visual sensory cues such as position, color, shape, etc. Previou...