Abstract--We describe some high-level approaches to estimating confidence scores for the words output by a speech recognizer. By "high-level" we mean that the proposed me...
Abstract--The notion of the necessary criticality (both with respect to path and to activity) of a network with imprecisely defined (by means of intervals or fuzzy intervals) activ...
Many problems in practically all fields of science, engineering and technology involve global optimization. It becomes more and more important to develop the efficient global opti...
This paper presents a hierarchical multi-state pose-dependent approach for facial feature detection and tracking under varying facial expression and face pose. For effective and e...
Singular Value Decomposition (and Principal Component Analysis) is one of the most widely used techniques for dimensionality reduction: successful and efficiently computable, it ...