Reuters tries facial recognition to track video content
The technology looks into video frames to identify the faces of people.
When searching for a person’s name, Viewdle’s technology automatically brings the user to the first frame in which that person appears. It also lists how long that person appears in that segment and other parts of the clip where that person is present.
According to journalism.co.uk’s observations, the technology isn’t perfectly fine tuned yet, but “is largely impressive and Viewdle says it's building the world's biggest people-in-video reference database.”
Reuters recently partnered with Attributor to be able to track its text content and pictures on the Web.
Source: Journalism.co.uk through European Journalism Centre
0 TrackBacks
Listed below are links to blogs that reference this entry: Reuters tries facial recognition to track video content.
TrackBack URL for this entry: http://www.editorsweblog.org/mt/mt-tb.cgi/1781

hallo. ich bin interessiert für die Frau das sagt immer alle für Sternen, ich bin interessiert für die magician Machts und ucha habe probleme mit einen boy hier, ich war sehr froh bis ich kennen gelernt hable