Pharos has announced co-development of the OpenFARM rich-media analysis framework. Initiated by the Computer Vision laboratory of Kingston University, OpenFARM is described as a bridge between data-based driven asset management systems and independently developed software.
“OpenFARM offers broadcasters freedom to add third-party software modules to their content management systems in essentially the same way that they select hardware for their central apparatus room,” explained Pharos commercial director Roger Heath. “It unites the worlds of signal-processing and media asset management to form a single market-place.”
Pharos has created two example plug-ins for the project. One detects video scene boundaries. The other converts programme credits into searchable text which is used to create metadata such as the names and titles of performers and production crew. It has also put an Open Farm interface into Mediator so the data from each plug-in becomes an accessible resource.
“The OpenFARM framework enables any given broadcast system to integrate any chosen analysis component,” said Dr James Orwell, Kingston University Academic Contributor and senior manager of the project. “This benefits broadcast users by allowing them to add to their tool set without significant disruption. It also creates opportunities for any given tool to be deployed more widely. OpenFARM is freely available to any company that wants to develop plug-ins or use third-party plug-ins in its own systems.”
An example plug-in signal processing algorithm, the framework, source code and the control centre can be downloaded for evaluation from www.openfarm.tv.