Archiving & Classification Tools for the Intelligent Video Ecosystem
The cost of manual metadata production is high, especially for audiovisual content, where a time-consuming inspection is usually required in order to identify the most appropriate annotations. There is a growing need from digital content industries for solutions capable of automating such a process.
ACTIVE is a platform for indexing and cataloging audiovisual collections through the automatic recognition of faces and speakers. The platform is totally accessible via browser and can be integrated with third-parties applications through its REST API. ACTIVE is based on a modular platform, which can be easily extended and adapted to a wide range of usage scenarios.
- The platform is able to distribute asynchronous jobs across several computers (workers), in order to reduce the execution time of CPU-intensive operations (e.g. transcoding, face detection/tracking/recognition). Workers can span over multiple computers, or multi-processor computers, or multi-core processors;
- a plug-in system provides a simple mechanism for extending the platform with server-side scripts.
ACTIVE is targeted both to end-users (mainly digital content industries) and developers who intend to build extra services and tools on top of ACTIVE API.
Digital preservation, digital libraries, video production, cultural heritage, new media, publishing, etc..
- Project Web site: http://active.crs4.it
- Maurizio Agelli, Maurizio Pintus, Felice Colucci, Alessandro Sassu, Federico Santamaria, Nicola Corona,
Digital Asset Management: Indicizzazione di contenuti audiovisuali mediante riconoscimento dei volti e degli speaker. Collana Seminari per la valorizzazione dei risultati della ricerca al CRS4, Cagliari, 13 Maggio 2015.
- Pintus Maurizio, Agelli Maurizio, Colucci Felice, Corona Nicola, Santamaria Federico, Sassu Alessandro - ACTIVE, an Extensible Cataloging Platform for Automatic Indexing of Audiovisual Content - Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications DOI: 0005722205740581 Volume: 134.