Automated multimedia content recognition
AssigneeMicrosoft Technology Licensing, LLC
InventorsEran Borenstein, Achiezer Brandt, Eitan Sharon, Mehmet Tek
Life of this patent
Estimating the term…
Abstract
An automated content recognition system accurately and reliably generates content identification information for multimedia content without accessing the multimedia content or a reliable source of the multimedia content. The system receives content-based queries having fingerprints of multimedia content. The system compares the individual queries to one another to match queries and thereby form query clusters that correspond to the same multimedia content. The system aggregates identification information from the queries in a cluster to generate reliable content identification information from otherwise unreliable identification information.
Claims20
A computer-readable storage medium having computer-executable instructions for programming a processor to perform a method, the method comprising:
- receiving from a plurality of client devices a plurality of queries for the identification of media content, each query including a media fingerprint;
- generating a plurality of query clusters based on comparisons of the media fingerprints of the plurality of queries, each query cluster including queries with one or more matching media fingerprints;
- for each query cluster, determining from at least a subset of queries in the query cluster textual data associated with the one or more matching media fingerprints;
- generating content identification information for each query cluster based on the textual data for the subset of queries;
- and
- generating a response to each query that includes the content identification information for a corresponding query cluster.
A computer-implemented method of automated content recognition, comprising:
- receiving from a plurality of clients a plurality of queries for the identification of media content, each query including a media fingerprint;
- generating from the plurality of queries query clusters based on comparisons of the media fingerprints of the plurality of queries;
- comparing the media fingerprint of at least one query from each query cluster to a base set of known media fingerprints;
- returning content identification information from the base set for each query cluster in which the at least one query matches a known media fingerprint of the base set;
- and
- for each query cluster in which the at least one query does not match a known media fingerprint,
- determining textual data associated with the media fingerprint from at least a subset of queries of the query cluster,
- generating content identification information based on the textual data for the subset of queries of the query cluster, and
- generating a response to each query of the cluster, the response including the generated content identification information for the query cluster.
A computing device, comprising:
- one or more storage devices comprising processor readable code and a catalog including media fingerprints and content identification information for media content;
- and
- one or more processors in communication with the one or more storage devices, the one or more processors execute the processor readable code to:
- receive from a plurality of clients over one or more networks a plurality of queries for the identification of media content, each query includes a media fingerprint corresponding to the media content,
- generate a plurality of query clusters, each query cluster including queries of the plurality that have a matching media fingerprint,
- determine whether one or more media fingerprints from each query cluster matches a media fingerprint from the catalog,
- for each query cluster that matches a media fingerprint from the catalog, generate content identification information for each query of the query cluster based on corresponding content identification information from the catalog, and
- for each query cluster that does match a media fingerprint from the catalog, generate content identification information for each query of the query cluster based on textual data associated with at least a subset of queries from the query cluster.
Dependencies are inferred from the claim text.