In this paper we present a novel approach for labeling clusters of multimedia content that leverages supervised classification techniques in conjunction with unsupervised cluster...
In extended video sequences, individual frames are grouped into shots which are defined as a sequence taken by a single camera, and related shots are grouped into scenes which are...
In this paper, a no reference bit stream model for quality assessment of SD and HD H.264/AVC video sequences based on packet loss visibility is proposed. The method considers the ...
Savvas Argyropoulos, Alexander Raake, Marie-Neige ...
This paper presents an approach that incorporates Canonical Correlation Analysis (CCA) for monocular 3D face pose and facial animation estimation. The CCA is used to find the depe...
Boosting is a remarkably simple and flexible classification algorithm with widespread applications in computer vision. However, the application of boosting to nonEuclidean, infini...