Scipediacontent: Scipediacontent moved page Draft Content 551843967 to Moisan et al 2011a

2021-02-03T16:36:20Z

Scipediacontent moved page Draft Content 551843967 to Moisan et al 2011a

Scipediacontent: Created page with " == Abstract == International audience; Abstract. Modernvisionsystemsmustrunincontinuallychangingcontexts.For example, a system to detect vandalism in train stations must fun..."

2021-02-03T16:36:17Z

Created page with " == Abstract == International audience; Abstract. Modernvisionsystemsmustrunincontinuallychangingcontexts.For example, a system to detect vandalism in train stations must fun..."

New page

== Abstract ==

International audience; Abstract. Modernvisionsystemsmustrunincontinuallychangingcontexts.For example, a system to detect vandalism in train stations must function during the day and at night. The vision components for acquisition and detection used dur- ing daytime may not be the same as those used at night. The system must adapt to a context by replacing running components such as image acquisition from color to infra-red. This adaptation must be dynamic with detection of context, decision on change in system configuration, followed by the seamless execution of the new configuration. All this must occur while minimizing the impact of dy- namic change on validity of detection and loss in performance. We present Girgit, a context-aware vision system for scene understanding, that dynamically orches- trates a set of components. A component encapsulates a vision-related algorithm such as from the OpenCV library. Girgit inherently provides loading/caching of multiple component instances, system reconfiguration, management of incoming events to suggest actions such as component re-configuration and replacement of components in pipelines. Given the surplus architectural layer for dynamic adap- tation one may ask, does Girgit degrade scene understanding performance? We perform several empirical evaluations on Girgit using metrics such as frame-rate and adaptation time to answer this question. For instance, the average adaptation time between change in configurations is less than 2 μs with caching, while 8 ms without caching. This in-turn has negligible effect on scene understanding perfor- mance with respect to static C++ implementations for most practical purposes.

== Original document ==

The different versions of the original document can be found in:

* [https://hal.inria.fr/inria-00616642 https://hal.inria.fr/inria-00616642]

* [http://link.springer.com/content/pdf/10.1007/978-3-642-23968-7_20 http://link.springer.com/content/pdf/10.1007/978-3-642-23968-7_20],
: [http://dx.doi.org/10.1007/978-3-642-23968-7_20 http://dx.doi.org/10.1007/978-3-642-23968-7_20] under the license http://www.springer.com/tdm

* [https://hal.inria.fr/inria-00616642 https://hal.inria.fr/inria-00616642],
: [https://link.springer.com/chapter/10.1007%2F978-3-642-23968-7_20 https://link.springer.com/chapter/10.1007%2F978-3-642-23968-7_20],
: [https://dblp.uni-trier.de/db/conf/icvs/icvs2011.html#RochaSMR11 https://dblp.uni-trier.de/db/conf/icvs/icvs2011.html#RochaSMR11],
: [https://rd.springer.com/chapter/10.1007/978-3-642-23968-7_20 https://rd.springer.com/chapter/10.1007/978-3-642-23968-7_20],
: [https://academic.microsoft.com/#/detail/100874462 https://academic.microsoft.com/#/detail/100874462]

← Older revision	Revision as of 16:36, 3 February 2021
(No difference)

Moisan et al 2011a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 551843967 to Moisan et al 2011a

Scipediacontent: Created page with " == Abstract == International audience; Abstract. Modernvisionsystemsmustrunincontinuallychangingcontexts.For example, a system to detect vandalism in train stations must fun..."