In an automated feature-based motion analysis of multiple frames, correspondence da ta are usually noisy and fragmented. We developed a technique that gradually refines the initial noisy correspondence d a t a and links fragments of a single trajectory using feedback from 3-0 motion estimation. First, 3-0 motion parameters are estimated using the initial correspondence data. Then, each noisy tr...

In this paper we propose a confidence measure for block-based motion vector field. The measure is calculated as an average of two a posteriori estimates which reflect various aspects of MVF accuracy: motion-compensated interframe difference distribution and motion vectors distribution. Experimental results show that the proposed measure outperforms its contemporary counterpart while demanding l...

This paper describes a new method for creating visually realistic moving facial image sequences that retain an actor's personality (individuality, expression and characteristic movements) while altering the facial appearance along a certain specified facial dimension. We combine two existing technologies, facial feature tracking and facial image transformation, to create the sequences. Examples...

The process of shot break detection is a fundamental component in automatic video indexing, editing and archiving. This paper introduces a novel approach to the detection and classification of shot transitions in video sequences including cuts, fades and dissolves. It uses the average interframe correlation coefficient and block-based motion estimation to track image blocks through the video se...