US 10,762,352 B2 · Method and system for the automatic identification of fuzzy copies of video content

Abstract

There is disclosed a method to automatically identify fuzzy video content copies, which is performed on a computer device that has access to a previously generated database of original video content. The method comprises: receiving identifiers for video content; extracting each Nth frame from the video stream of received video content to obtain a sequence of frames, where N>1; identifying the sequence of scenes by comparing at least fragments of adjacent frames; creating a vector of the sequence of scenes of the received video content; comparing the vector of the sequence of scenes of received video content with the vector of the sequence of scenes of at least one template of original video content from a previously generated database of original video content; identifying a fuzzy copy of original video content when the sequence of scenes of original video content and scenes of received video content coincide to an extent that exceeds a predefined threshold.

Claims18

Dependent claims

A method for automatically identifying a fuzzy video content copy, the method executable on a computer device that has access to an original content database;

the method comprising:
receiving an identifier of video content in a form of a video stream;
extracting each Nth frame from the video stream to obtain a sequence of frames, where N is greater than one;
identifying a sequence of scenes by comparing at least fragments of adjacent frames from the sequence of frames;
creating, based on the sequence of scenes, a scene sequence vector;
obtaining, from the original content database, a template vector representative of an original sequence of scenes of at least one template of an original video content;
comparing the scene sequence vector with the template vector;
in response to the comparing identifying that the scene sequence vector and the template vector overlap to a degree that exceeds a predefined threshold, identifying the video content as the fuzzy video content copy of the original video content.

5 more elements

A computer device configured to automatically identify fuzzy copies of video content, the computer device having access to a original video content database, the computer devices being connectable to a data transmission network;

the computer device including a communication interface and at least one computer processor that is functionally connected to the communication interface, the at least one computer processor being configured to:
receive an identifier of video content in a form of a video stream;
extract each Nth frame from the video stream to obtain a sequence of frames, where N is greater than one;
identify a sequence of scenes by comparing at least fragments of adjacent frames from the sequence of frames;
create, based on the sequence of scenes, a scene sequence vector;
obtain, from the original content database, a template vector representative of an original sequence of scenes of at least one template of an original video content;
compare the scene sequence vector with the template vector;
in response to the comparing identifying that the scene sequence vector and the template vector overlap to a degree that exceeds a predefined threshold, identify the video content as the fuzzy video content copy of the original video content.

5 more elements

Dependencies are inferred from the claim text.

Method and system for the automatic identification of fuzzy copies of video content

Life of this patent

Method and system for the automatic identification of fuzzy copies of video content

Life of this patent

Abstract

Claims18