Learning visual models of object categories notoriously requires thousands of training examples; this is due to the diversity and richness of object appearance which requires mode...
In this paper, we take a predefined geometry shape as a constraint for accurate shape alignment. A shape model is divided in two parts: fixed shape and active shape. The fixed shap...
In this paper, we focus on the challenge that users face in processing messages on the web posted in participatory media settings, such as blogs. It is desirable to recommend to us...
In this paper a method for extraction of mid-level semantics from sign language videos is proposed, by employing high level domain knowledge. The semantics concern labeling of the ...
Decision makers (humans or software agents alike) are faced with the challenge of examining large volumes of information originating from heterogeneous sources with the goal of as...