Towards computer vision with description logics: Some recent progress

A description logic (DL) is a knowledge representation formalism which may provide interesting inference services for diverse application areas. This paper first gives an overview of the benefits which a DL may provide for Computer Vision. The main body of the paper presents recent work at Hamburg University on extending DLs to handle spatial reasoning and default reasoning. 1: Why is Description Logic Interesting for Computer Vision? This contribution discusses the merits of description logics (DLs) for Computer Vision (CV) and reports about some recent work on DL extensions at Hamburg University. Our goal is to make DLs more useful for diverse applications, in particular those involving concrete real-life phenomena which play a part in diagnosis, configuration and – last not least – in CV. This work extends results previously published in [9]. DL is the family name of object-based knowledge-representation formalisms in the spirit of KL-ONE which have been introduced 20 years ago [4] with the main purpose of providing formal semantics for semantic nets, thus providing the logical foundations for knowledge-based inferences. A particular DL typically realizes a particular subset of First Order Predicate Logic (FOPL). Muc