System and method for generating a multimedia summary of multimedia streams
First Claim
1. A method for producing a multimedia summary of at least one multimedia stream the summary comprising key elements selected from said at least one multimedia stream, the method comprising:
- a.) one of receiving and retrieving said at least one multimedia stream comprising video, audio and text information;
b.) dividing the at least one multimedia stream into a video substream, an audio sub-stream and a text sub-stream;
c.) identifying for potential inclusion in the summary video, audio and text key elements from said video, audio and text sub-streams, respectively;
d.) computing an importance value for the identified video, audio and text key elements identified at said step (c);
e.) first filtering the identified video, audio and text key elements to exclude those key elements whose associated importance value is less than a pre-defined video, audio and text importance threshold, respectively; and
f.) second filtering the remaining key elements from said step (e) in accordance with a user profile;
g.) third filtering the remaining key elements from said step (f) in accordance with network and user device constraints; and
h.) outputting a multimedia summary which comprises key elements remaining from said step (g).
6 Assignments
0 Petitions
Accused Products
Abstract
A system facilitates and enhances review of one or more multimedia input streams that includes some combination of video, audio and text information, generating a multimedia summary, thereby enabling a user to better browse and/or decide on viewing the multimedia input streams in their entirety. The multimedia summary is constructed automatically, based in part on system specifications, user specifications and network and device constraints. In a particular application of the invention, the input multimedia streams represent news broadcasts (e.g., television news program, video vault footage). In such a particular application, the invention can enable the user to automatically receive a summary of the news stream in accordance with previously provided user preferences and in accordance with prevailing network and user device constraints.
37 Citations
27 Claims
-
1. A method for producing a multimedia summary of at least one multimedia stream the summary comprising key elements selected from said at least one multimedia stream, the method comprising:
-
a.) one of receiving and retrieving said at least one multimedia stream comprising video, audio and text information; b.) dividing the at least one multimedia stream into a video substream, an audio sub-stream and a text sub-stream; c.) identifying for potential inclusion in the summary video, audio and text key elements from said video, audio and text sub-streams, respectively; d.) computing an importance value for the identified video, audio and text key elements identified at said step (c); e.) first filtering the identified video, audio and text key elements to exclude those key elements whose associated importance value is less than a pre-defined video, audio and text importance threshold, respectively; and f.) second filtering the remaining key elements from said step (e) in accordance with a user profile; g.) third filtering the remaining key elements from said step (f) in accordance with network and user device constraints; and h.) outputting a multimedia summary which comprises key elements remaining from said step (g). - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
-
19. A system for producing a multimedia summary of at least one multimedia stream, the summary comprising key elements selected from said at last one multimedia stream, the system comprising:
- a modality recognition and division (MRAD) module comprising a story segment identifier (SSI) module, an audio identifier (AI) module and a text identifier (TI) module, the MRAD module communicatively coupled to a first external source for receiving said at least one multimedia stream, the MRAD module communicatively coupled to a second external source for receiving said at least one multimedia stream, the MRAD module dividing said at least one multimedia stream into a video, an audio and a text sub-stream and outputting said video, audio and text sub-streams to a key element identifier (KEI) module, the KEI module comprising a feature extraction (FE) module and an importance value (IV) module for identifying key elements from within said video, audio and text sub-streams and assigning importance values thereto, the KEI module communicatively coupled to a key element filter (KEF) for receiving the identified key elements and filtering said key elements that exceed a pre-determined threshold criteria, the KEF module communicatively coupled to a user profile filter (UPF) for receiving filtered key elements and further filtering said filtered key elements in accordance with a user profile, the UPF module communicatively coupled to a network and device constraint (NADC) module, said NADC module receiving said further filtered key elements and further filtering said further filtered key elements in accordance with network and/or user device constraints, the NADC module outputting a multimedia summary of said at least one multimedia stream which comprises key elements remaining after said further filtering in the NADC module.
- View Dependent Claims (20, 21, 22, 23, 24, 25)
-
26. A non-transitory computer readable medium having computer readable code means embodied thereon, said computer readable program code means comprising:
- an act of one of receiving and retrieving said at least one multimedia stream comprising video, audio and text information;
an act of dividing said at least one multimedia stream into a video sub-stream, an audio sub-stream and a text sub-stream an act of identifying for potential inclusion in the summary video, audio and text key elements from said video, audio and text sub-streams, respectively;
an act of computing an importance value for the identified video, audio and text key elements identified at said identification act;
an act of first filtering the identified video, audio and text key elements to exclude those key elements whose associated importance value is less than a pre-defined video, audio and text importance threshold, respectively; and
an act of second filtering the remaining key elements from said first filtering act in accordance with a user profile;
an act of third filtering the remaining key elements from said second filtering act in accordance with network and user device constraints; and
an act of outputting a multimedia summary which comprises key elements remaining from said third filtering act. - View Dependent Claims (27)
- an act of one of receiving and retrieving said at least one multimedia stream comprising video, audio and text information;
Specification