Media content identification on mobile devices
A mobile device responds in real time to media content presented on a media device, such as a television. The mobile device captures temporal fragments of audio-video content on its microphone, camera, or both and generates corresponding audio-video query fingerprints. The query fingerprints are transmitted to a search server located remotely or used with a search function on the mobile device for content search and identification. Audio features are extracted and audio signal global onset detection is used for input audio frame alignment. Additional audio feature signatures are generated from local audio frame onsets, audio frame frequency domain entropy, and maximum change in the spectral coefficients. Video frames are analyzed to find a television screen in the frames, and a detected active television quadrilateral is used to generate video fingerprints to be combined with audio fingerprints for more reliable content identification.
1. A computer-implemented method comprising:
fingerprinting, by a client media device, portions of a media sequence that is presented during a period of time by the client media device so as to obtain reference fingerprints;
storing, by the client media device, the reference fingerprints in a reference database of the client media device, wherein the reference database comprises a revolving buffer of programmable length;
after storing the reference fingerprints, receiving, by the client media device from a server device, a query fingerprint corresponding to television programming scheduled to be presented during the period of time, wherein the television programming has a duration, and wherein the reference fingerprints are stored in the revolving buffer for at least the duration of the television programming and then subsequently removed from the revolving buffer;
in response to receiving the query fingerprint, searching, by the client media device, the reference database for any reference fingerprints that match the query fingerprint; and
based on identifying a reference fingerprint of the reference database that matches the query fingerprint, initiating an actionable event on the client media device.
2. The computer-implemented method of claim 1 , wherein the method further comprises refreshing, by the client media device, the reference database according to a programmable schedule.
3. The computer-implemented method of claim 1 , wherein refreshing the reference database comprises:
removing an oldest reference fingerprint from the reference database; and
adding a new reference fingerprint to the reference database.
4. The computer-implemented method of claim 1 , further comprising:
determining, by the client media device, that the reference database is full; and
based on the determining that the reference database is full, removing, by the client media device, an oldest reference fingerprint from the reference database.
5. The computer-implemented method of claim 1 , wherein initiating the actionable event comprises replacing a generic advertisement with a targeted advertisement.
6. The computer-implemented method of claim 1 , wherein initiating the actionable event comprises providing a menu of enhancement programming options on the client media device.
7. The computer-implemented method of claim 6 , wherein the enhancement programming options are related to the television programming.
8. The computer-implemented method of claim 1 , wherein initiating the actionable event comprises sending, to a server device, an indication that the client media device presented the television programming.
9. The computer-implemented method of claim 1 , wherein the client media device is a network-connected client video device, and wherein the reference fingerprints comprise video fingerprints.
10. The computer-implemented method of claim 1 , wherein the client media device is a network-connected client audio device, and wherein the reference fingerprints comprise audio fingerprints.
11. A client media device comprising:
one or more processors; and
a computer-readable medium having stored therein instructions that are executable by the one or more processors to cause the client media device to perform functions comprising:
fingerprinting portions of a media sequence that is presented during a period of time by the client media device so as to obtain reference fingerprints,
storing the reference fingerprints in a reference database of the client media device, wherein the reference database comprises a revolving buffer of programmable length,
after storing the reference fingerprints, receiving, from a server device, a query fingerprint corresponding to television programming scheduled to be presented during the period of time, wherein the television programming has a duration, and wherein the reference fingerprints are stored in the revolving buffer for at least the duration of the television programming and then subsequently removed from the revolving buffer,
in response to receiving the query fingerprint, searching the reference database for any reference fingerprints that match the query fingerprint, and
based on identifying a reference fingerprint of the reference database that matches the query fingerprint, initiating an actionable event on the client media device.
12. The client media device of claim 11 , wherein the functions further comprise refreshing the reference database according to a programmable schedule.
13. The client media device of claim 11 , wherein the functions further comprise:
determining that the reference database is full; and
based on the determining that the reference database is full, removing an oldest reference fingerprint from the reference database.
14. The client media device of claim 11 , wherein initiating the actionable event comprises replacing a generic advertisement with a targeted advertisement.
15. The client media device of claim 11 , wherein initiating the actionable event comprises providing a menu of enhancement programming options on the client media device.
16. The client media device of claim 11 , wherein initiating the actionable event comprises sending, to a server device, an indication that the client media device presented the television programming.
17. A non-transitory computer-readable medium having stored therein instructions that are executable by one or more processors to cause a client media device to perform functions comprising:
fingerprinting portions of a media sequence that is presented during a period of time by the client media device so as to obtain reference fingerprints;
storing the reference fingerprints in a reference database of the client media device, wherein the reference database comprises a revolving buffer of programmable length;
after storing the reference fingerprints, receiving, from a server device, a query fingerprint corresponding to television programming scheduled to be presented during the period of time, wherein the television programming has a duration, and wherein the reference fingerprints are stored in the revolving buffer for at least the duration of the television programming and then subsequently removed from the revolving buffer;
in response to receiving the query fingerprint, searching the reference database for any reference fingerprints that match the query fingerprint; and
based on identifying a reference fingerprint of the reference database that matches the query fingerprint, initiating an actionable event on the client media device.
18. The non-transitory computer-readable medium of claim 17 , wherein the functions further comprise refreshing the reference database according to a programmable schedule.
19. The non-transitory computer-readable medium of claim 17 , wherein initiating the actionable event comprises replacing a generic advertisement with a targeted advertisement.
20. The non-transitory computer-readable medium of claim 17 , wherein initiating the actionable event comprises sending, to a server device, an indication that the client media device presented the television programming.