Methods for using extracted feature vectors to perform an action associated with a work identifier
A method comprising the steps of: receiving, from a user electronic device at a computer system, one or more extracted feature vectors that are compact representations of one or more video frames of a video displayed at the user device; identifying, at the computer system, one or more products related to the one or more video frames based at least in part on an approximate nearest neighbor search of an electronic database of reference feature vectors corresponding to one or more reference works that identifies any match within a threshold of at least one of the reference feature vectors to the one or more extracted feature vectors but does not guarantee to identify the closest match to the one or more extracted feature vectors; providing, by the computer system to the user electronic device, the identified one or more products along with action information associated with the identified one or more products, wherein the action information is related to an action to be performed at the user electronic device; receiving, at the computer system from the user electronic device, a selection of one of the identified one or more products; and providing, to the electronic device, machine readable instructions for performance of the action associated with the selected one of the identified one or more products.
1. A computer implemented method comprising:
(a) obtaining, by a first computer system, a first electronic media work, wherein the first electronic media work comprises at least a first audio file;
(b) assigning, by the first computer system, a work identifier to the first electronic media work;
(c) extracting, by the first computer system, from at least a portion of the at least the first audio file, using, at least in part, a frequency-based decomposition of the portion produced by a Fourier frequency decomposition, a first electronic feature vector comprising a first compact representation of the first electronic media work;
(d) sending, from the first computer system to a second computer system, the first electronic feature vector and the work identifier;
(e) receiving, by the first computer system from the second computer system, first action information and the work identifier, wherein:
(1) the second computer system is operatively connected to a database comprising:
(A) first electronic data related to identification of one or more reference electronic works; and
(B) second electronic data related to action information comprising an action to perform corresponding to each of the one or more reference electronic works; and
(2) the first action information is determined by the second computer system by using a sub-linear search of the first electronic data that identifies a match to the first electronic feature vector within a threshold but does not guarantee to identify the closest match to the first electronic feature vector, wherein the first electronic data is organized using a clustering technique to eliminate a subset of the first electronic data to be searched by performing a comparison of cluster features;
(f) associating, by the first computer system, the first action information with the work identifier associated with the first electronic media work; and
(g) performing, by the first computer system, the first action information in association with the first electronic media work.
2. The method of claim 1 , wherein the first action information comprises information related to a URL.
3. The method of claim 1 , wherein an action associated with the first action information further comprises redirecting a request to an associate website by a network address translation.
4. The method of claim 1 , wherein an action associated with the first action information further comprises automatically dialing a telephone number.
5. The method of claim 1 , wherein the first electronic media work further comprises at least a first video file.
6. The method of claim 1 , wherein the first action information is determined by the second computer system further based on a temporal sequence of the extracted first electronic feature vector.
7. The method of claim 1 , further comprising the steps of:
(h) generating, at the first computer system, machine-readable instructions to be executed by a first electronic media device in performing the determined first action information; and
(i) providing, from the first computer system to the first electronic media device, the machine-readable instructions in response to a query received from the first electronic media device related to the first electronic media work.
8. The method of claim 7 , wherein the machine-readable instructions are further based on device information relating to the first electronic media device at which the first electronic media work is rendered.
9. The method of claim 8 , wherein the device information comprises a geographic location of the first electronic media device.
10. The method of claim 1 , further comprising the steps of:
(h) obtaining, by the first computer system, a second electronic media work, wherein the second electronic media work comprises at least a second audio file;
(i) assigning, by the first computer system, a second work identifier to the second electronic media work;
(j) extracting, by the first computer system, from at least a second portion of the at least the second audio file, using, at least in part, the frequency-based decomposition of the second portion produced by a Fourier frequency decomposition, a second electronic feature vector comprising a second compact representation of the second electronic media work;
(k) sending, from the first computer system to a second computer system, the second electronic feature vector and the second work identifier;
(l) receiving, by the first computer system from the second computer system, second action information, wherein:
(1) the second computer system is operatively connected to the database; and
(2) the second action information is determined by the second computer system by using a sub-linear search of the first electronic data that identifies a match to the second electronic feature vector within a threshold but does not guarantee to identify the closest match to the first electronic feature vector, wherein the first electronic data is organized using a clustering technique to eliminate a subset of the first electronic data to be searched by performing a comparison of cluster features and wherein none of the one or more reference electronic works is identified to match the second electronic feature vector;
(m) associating, by the first computer system, the second action information with the second work identifier associated with the second electronic media work; and
(n) performing, by the first computer system, the second action information in association with the second electronic media work.