"The paper deals with a solution for visual surveillance metadata management. Data coming from many cameras is annotated using computer vision units to produce metadata representing moving objects in their states. It is assumed that the data is often uncertain, noisy and some states are missing.

The solution consists of the following three layers: (a) data cleaning layer - improves quality of the data by smoothing it and by filling in missing states in short sequences referred to as tracks that represent a composite state of a moving object in a spatiotemporal subspace followed by one camera. (b) Data integration layer - assigns a global identity to tracks that represent the same object. (c) Persistence layer - manages the metadata in a database so that it can be used for online identification and offline querying, analyzing and mining. A Kalman filter technique is used to solve (a) and a classification based on the moving object's state and its visual properties is used in (b). An object model for"@en . . . "458027" . "The paper deals with a solution for visual surveillance metadata management. Data coming from many cameras is annotated using computer vision units to produce metadata representing moving objects in their states. It is assumed that the data is often uncertain, noisy and some states are missing.

The solution consists of the following three layers: (a) data cleaning layer - improves quality of the data by smoothing it and by filling in missing states in short sequences referred to as tracks that represent a composite state of a moving object in a spatiotemporal subspace followed by one camera. (b) Data integration layer - assigns a global identity to tracks that represent the same object. (c) Persistence layer - manages the metadata in a database so that it can be used for online identification and offline querying, analyzing and mining. A Kalman filter technique is used to solve (a) and a classification based on the moving object's state and its visual properties is used in (b). An object model for" . . . "Spr\u00E1va metadat pro vizu\u00E1ln\u00ED dohled"@cs . . "RIV/00216305:26230/07:PU70857!RIV08-MSM-26230___" . . "Visual Surveillance Metadata Management" . "RIV/00216305:26230/07:PU70857" . . . "Eighteenth International Workshop on Database and Expert Systems Applications" . . . "2"^^ . "79-83" . "[EA4552812915]" . "978-0-7695-2932-5" . . . . "2"^^ . . . "Regensburg" . . "Regensburg" . "Spr\u00E1va metadat pro vizu\u00E1ln\u00ED dohled"@cs . . "Visual Surveillance Metadata Management" . . . "5"^^ . "2007-09-03+02:00"^^ . . . "Visual Surveillance Metadata Management"@en . . "Z(MSM0021630528)" . "\u010Cl\u00E1nek se zab\u00FDv\u00E1 \u0159e\u0161en\u00EDm spr\u00E1vy metadat pro vizu\u00E1ln\u00ED dohled. Data p\u0159ich\u00E1zej\u00EDc\u00ED z mnoha kamer jsou anotov\u00E1na pomoc\u00ED jednotek po\u010D\u00EDta\u010Dov\u00E9ho vid\u011Bn\u00ED, kter\u00E9 produkuj\u00ED metadata reprezentuj\u00EDc\u00ED pohybuj\u00EDc\u00ED objekty v jejich stavech. P\u0159edpokl\u00E1d\u00E1me, \u017Ee data jsou nespolehliv\u00E1, za\u0161um\u011Bn\u00E1 a n\u011Bkter\u00E9 stavy chyb\u011Bj\u00ED.
\u0158e\u0161en\u00ED spo\u010D\u00EDv\u00E1 ve t\u0159ech vrstv\u00E1ch: (a) \u010Di\u0161t\u011Bn\u00ED dat - zvy\u0161uje jejich kvalitu (pomoc\u00ED Kalmanova filtu), (b) integrace - p\u0159i\u0159azuje pohybuj\u00EDc\u00EDm se objekt\u016Fm glob\u00E1ln\u00ED identifik\u00E1tor (Bayesovk\u00E1 klasifikace, SVM), (c) persistentn\u00ED vrstva zaji\u0161\u0165uje spr\u00E1vu metadat, dotazov\u00E1n\u00ED v re\u00E1ln\u00E9m \u010Dase, anal\u00FDzu dat a umo\u017E\u0148uje dolov\u00E1n\u00ED.

"@cs . "26230" . . "Zendulka, Jaroslav" . "Visual surveillance, metadata management, cameras, vision units, moving objects, data cleaning, integration, persistence, Kalman filter, classification, object model."@en . . . "IEEE Computer Society Press" . "Visual Surveillance Metadata Management"@en . "Chmela\u0159, Petr" .