The objective of active recognition is to iteratively collect the next "best" measurements (e.g., camera angles or viewpoints), to maximally reduce ambiguities in recogn...
The Conditional Random Fields (CRF) model, using
patch-based classification bound with context information,
has recently been widely adopted for image segmentation/
labeling. In...
RDF (core Semantic Web standard) is not originally appropriate for context representation, because of its initial focus on the ordinary Web resources, such as web pages, files, dat...
Sergiy Nikitin, Vagan Y. Terziyan, Yaroslav Tsaruk...
Abstract. This paper proposes a new approach to learning a discriminative model of object classes, incorporating appearance, shape and context information efficiently. The learned ...
Jamie Shotton, John M. Winn, Carsten Rother, Anton...
The context of the user queries, preceding a given query, is utilized to improve the effectiveness of query classification. Earlier efforts utilize fixed number of preceding queri...