Media fingerprinting and identification system
The overall architecture and details of a scalable video fingerprinting and identification system that is robust with respect to many classes of video distortions is described. In this system, a fingerprint for a piece of multimedia content is composed of a number of compact signatures, along with traversal hash signatures and associated metadata. Numerical descriptors are generated for features found in a multimedia clip, signatures are generated from these descriptors, and a reference signature database is constructed from these signatures. Query signatures are also generated for a query multimedia clip. These query signatures are searched against the reference database using a fast similarity search procedure, to produce a candidate list of matching signatures. This candidate list is further analyzed to find the most likely reference matches. Signature correlation is performed between the likely reference matches and the query clip to improve detection accuracy.
1. A method comprising:
obtaining, by one or more computing devices, a signature corresponding to a sequence of media content;
determining, by the one or more computing devices, a value representing a uniqueness of the signature relative to a set of other signatures, wherein the set of other signatures are derived from other media content;
determining, by the one or more computing devices, that the value representing the uniqueness of the signature satisfies a threshold; and
based at least on the determining that the value representing the uniqueness of the signature satisfies the threshold, entering, by the one or more computing devices, the signature in a media reference database.
2. The method of claim 1 , wherein the value representing the uniqueness of the signature is a term frequency, and wherein the term frequency is an occurrence within the set of other signatures of signatures that are within a threshold distance of the signature.
3. The method of claim 2 , wherein determining the value representing the uniqueness of the signature comprises determining the signatures that are within the threshold distance of the signature by comparing the signatures with the signature.
4. The method of claim 1 , wherein the entering of the signature in the media reference database is further based on a total number of signatures obtained for the sequence of media content.
5. The method of claim 4 , wherein the entering of the signature in the media reference database is further based on a control parameter that specifies a desired number of signatures per sequence of media content.
6. The method of claim 4 , further comprising obtaining one or more other signatures corresponding to the sequence of media content.
7. The method of claim 1 , wherein the entering of the signature in the media reference database is further based on a priority of the sequence of media content.
8. The method of claim 1 , wherein entering the signature in the media reference database comprises entering the signature using a hash index.
9. The method of claim 1 , wherein the sequence of media content is a reference video, and wherein the method further comprises storing, in the media reference database, the signature in association with video data for the reference video.
10. A non-transitory computer-readable medium having stored therein instructions that are executable to cause a computing system to perform functions comprising:
obtaining a signature corresponding to a sequence of media content;
determining a value representing a uniqueness of the signature relative to a set of other signatures, wherein the set of other signatures are derived from other media content;
determining that the value representing the uniqueness of the signature satisfies a threshold; and
based at least on the determining that the value representing the uniqueness of the signature satisfies the threshold, entering the signature in a media reference database.
11. The non-transitory computer-readable medium of claim 10 , wherein the value representing the uniqueness of the signature is a term frequency, and wherein the term frequency is an occurrence within the set of other signatures of signatures that are within a threshold distance of the signature.
12. The non-transitory computer-readable medium of claim 10 , wherein the entering of the signature in the media reference database is further based on a total number of signatures obtained for the sequence of media content.
13. The non-transitory computer-readable medium of claim 10 , wherein the entering of the signature in the media reference database is further based on a priority of the sequence of media content.
14. The non-transitory computer-readable medium of claim 10 , wherein entering the signature in the media reference database comprises entering the signature using a hash index.
15. The non-transitory computer-readable medium of claim 10 , wherein the sequence of media content is a reference video, and wherein the functions further comprise storing, in the media reference database, the signature in association with video data for the reference video.
16. A computing system comprising:
a processing unit; and
a non-transitory computer-readable medium having stored therein instructions that are executable to cause the computing system to perform functions comprising:
obtaining a signature corresponding to a sequence of media content,
determining a value representing a uniqueness of the signature relative to a set of other signatures, wherein the set of other signatures are derived from other media content,
determining that the value representing the uniqueness of the signature satisfies a threshold, and
based at least on the determining that the value representing uniqueness of the signature satisfies the threshold, entering the signature in a media reference database.
17. The computing system of claim 16 , wherein the value representing the uniqueness of the signature is a term frequency, and wherein the term frequency is an occurrence within the set of other signatures of signatures that are within a threshold distance of the signature.
18. The computing system of claim 16 , wherein the entering of the signature in the media reference database is further based on a total number of signatures obtained for the sequence of media content.
19. The computing system of claim 16 , wherein entering the signature in the media reference database comprises entering the signature using a hash index.
20. The computing system of claim 16 , wherein the sequence of media content is a reference video, and wherein the functions further comprise storing, in the media reference database, the signature in association with video data for the reference video.