information retrieval perspectives: information retrieval • • • • • • • • application(s) - surveillance psychological - information overload experimental - search engines algorithmic - (media) similarity metrics system - unified (media) repository presentation - user profile / integration search - precision & recall commercial - applications & certification media types text, images, audio, video, structured data retrieval image query -- all images with this person audio query -- identity of speaker text query -- all transactions with BANK Inc. video query -- all segments with victim complex queries -- convicted murderers with BANK transactions heterogeneous queries -- photograph + murderer + transaction complex heterogeneous queries -- in contact with + murderer + transaction precision -- how many answers are correct recall -- how many of the right documents are returned precision = ( returned and relevant ) / returned recall = ( returned and relevant ) / relevant frequency table term/document d0 snacks 1 0 0 drinks 1 0 3 rock-roll 0 1 d1 d2 1 indexing video content video v, frame f f has associated objects and activities objects and activities have properties object instance: (oid,os,ip) object-id -- oid object-schema -- os = (fd,fi) set of statements -- ip: name = v and name = v IN f video frame objects frame-dependent properties 1 Jane has(briefcase), at(path) - house door(closed) - briefcase 2 Jane - Dennis at(door) - house door(open) - briefcase has(briefcase), at(door) object frame-independent properties value Jane 35 age height 170cm house address ... color briefcase size brown color black 40 x 31 query language for video libraries segment retrievals -- exchange of briefcase object retrievals -- all people in v:[s,e] activity retrieval -- all activities in v:[s,e] property-based -- find all videos with object oid SQL, XML, logic-based, ... media abstractions media abstraction state -- smallest chunk of media data feature -- any object in a state attributes -- characteristics of objects feature extraction map -- to identify content relations -- to capture state-dependent information (inter)relations between 'states' or chunks VO = { (O_i,Q_i,C_i) | 1 <= i <= k } where C_1,...,C_k -- mutually exclusive conditions Q_1,...,Q_k -- queries O_1,...,O_k -- objects virtual objects user interface augmented virtuality information retrieval: issues • retrieval -- indexing & success metrics • video -- (temporal) description logic • applications -digital dossier Create a VR that realizes a digital dossier for a work of a particular artist. A digital dossier represents the information that is available for a particular work of art, or a collection of works, of a particular artist. http://www.cs.vu.nl/~eliens/casus/ perspectives: information retrieval • • • • • • • • application(s) - digital dossier psychological - focus experimental - user interaction algorithmic - (information) access system - unified presentation space presentation - embodied agents search - semantic annotation commercial - future systems