|
Philips' Video Fingerprinting system translates characteristic features extracted from the video file into a bit string called a fingerprint. Content is identified by comparing fingerprints of a video file received, for example over a peer-to-peer network, with the fingerprints of previously stored original video files in the database.
Philips' Video fingerprinting system enables users to automatically identify video by comparing unique 'fingerprints' extracted from the video with fingerprints in a specially compiled database. Fingerprint extraction occurs in a lightweight, client-side library.The database runs on a central server and comprises an advanced proprietary fingerprint search algorithm plus a database containing video fingerprints and metadata (identification data).
|