Naturalistic data sets for image and behavior analysis - 'normal' versus 'anomalous' events

Autor:	Philippe Rolland, William Krebs, Annetta Burger
Rok vydání:	2011
Předmět:	business.industry Computer science computer.software_genre Visualization Image (mathematics) Reference data Server Global Positioning System Computer vision Data mining Artificial intelligence Situational ethics business computer
Zdroj:	AVSS
DOI:	10.1109/avss.2011.6027345
Popis:	The approach and rationale used to create controlled data sets encompassing video, track, and reference data and meta-data is described in this paper. The data sets are designed to support the development of automated technologies for the detection of situational and operational behaviors of interest. A custom tool for the concurrent visualization and access of the data is also introduced.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::025d49b8bf72bafcc42ae0ff774d47fa https://doi.org/10.1109/avss.2011.6027345 Zobrazit plný text záznamu