Distributed and tiered architecture for content search and content monitoring
An efficient large scale search system for video and multi-media content using a distributed database and search, and tiered search servers is described. Selected content is stored at the distributed local database and tier1 search server(s). Content matching frequent queries, and frequent unidentified queries are cached at various levels in the search system. Content is classified using feature descriptors and geographical aspects, at feature level and in time segments. Queries not identified at clients and tier1 search server(s) are queried against tier2 or lower search server(s). Search servers use classification and geographical partitioning to reduce search cost. Methods for content tracking and local content searching are executed on clients. The client performs local search, monitoring and/or tracking of the query content with the reference content and local search with a database of reference fingerprints. This shifts the content search workload from central servers to the distributed monitoring clients.
1. A method for content searching, the method comprising:
generating fingerprints, in a user device, for a unit of multimedia content;
searching for the generated fingerprints in a found query (FQ) cache and in an unidentified query (UIQ) cache;
searching in a remote reference database if the generated fingerprints are not found in the FQ cache or in the UIQ cache; and
storing the unit of multimedia content and the generated fingerprints in the UIQ cache upon not finding the generated fingerprints in the remote reference database, wherein subsequent searches of the unit of multimedia content are found in the UIQ cache without searching the remote reference database.
2. The method of claim 1 further comprising: organizing the FQ cache and UIQ cache as a first tier database and the remote reference database as a second tier database.
3. The method of claim 1 further comprising:
searching in the FQ cache for the generated fingerprints and returning to the user device a menu of options if a match is found in the FQ cache; and
searching in the UIQ cache for the generated fingerprints and returning to the user device a not identified if a match is found in the UIQ cache.
4. The method of claim 3 further comprising:
requesting the user to send information in response to the match indicating the unit of multimedia content is in the UIQ cache including a program name and program source of the content associated with the query to be stored as metadata in the UIQ cache.
5. The method of claim 1 further comprising:
updating a count of repeated queries for the unit of multimedia content that is found in the UIQ cache, wherein the count is stored in the UIQ cache with metadata stored with the unit of multimedia content that is unidentified.
6. The method of claim 5 further comprising:
aging out the count of repeated queries for the unit of multimedia content found in the UIQ cache according to a reduced number of query searches for the unit of multimedia content.
7. The method of claim 1 further comprising:
updating counts of recent matches to generated fingerprints for separate units of multimedia content that are cached in the UIC cache, wherein the counts are used to determine which cached content is to be removed.
8. The method of claim 1 further comprising:
storing a rate of querying for matching fingerprints found in the remote reference database;
comparing the rate of querying with an established threshold to determine that the matching fingerprints represents popular content; and
storing the matching fingerprints in the FQ cache for quick access.
9. The method of claim 1 , wherein reference multimedia content is stored in the remote reference database in categories according to program type including sports, local news, and national news programs.
10. A method to reduce query search cost, the method comprising:
generating, in a user device, fingerprints for media content presented by a media presentation device;
searching for the generated fingerprints in a local reference database on the media device in response to a search query requested by a client, wherein the local reference database is a subset of a remote reference database;
searching in a local not identified cache on the media device if the generated fingerprints are not found in the local reference database; and
responding to the client with metadata regarding the media content upon finding the generated fingerprints in the local not identified cache.
11. The method of claim 10 further comprising:
searching the remote reference database if the generated fingerprints are not found in the local not identified cache; and
storing the media content and the generated fingerprints in the local not identified cache upon not finding the generated fingerprints in the remote reference database, wherein subsequent searches for the media content are found in the local not identified cache without searching the remote reference database.
12. The method of claim 10 further comprising:
searching the remote reference database if the generated fingerprints are not found in the local not identified cache; and
storing the media content and the generated fingerprints in the local reference database upon finding the generated fingerprints in the remote reference database, wherein subsequent searches for the media content are found in the local reference database without searching the remote reference database.
13. The method of claim 12 , wherein metadata including program name and program source obtained from the remote reference database are stored with the media content and the generated fingerprints in the local reference database.
14. The method of claim 10 further comprising:
tracking the media content by finding matches to the generated fingerprints in the local reference database without querying the remote reference database.
15. The method of claim 10 further comprising:
tracking the media content by finding matches to the generated fingerprint in the local not identified cache without querying the remote reference database.
16. The method of claim 10 further comprising:
maintaining the local not identified cache by using a least recently used caching strategy to keep the local not identified cache under a predetermined limit.
17. The method of claim 10 further comprising:
responding to the client, upon not finding the generated fingerprints in the local not identified cache, with a request for the client to provide the metadata regarding the media content including program name and program source for storage with the media content and the generated fingerprints in the local not identified cache.
18. The method of claim 17 , wherein the metadata includes a uniform resource locator (url) for the named program.
19. A computer readable non-transitory medium encoded with computer readable program data and code, the computer readable program data and code when executed perform:
generating, in a user device, fingerprints for media content presented by a media presentation device;
searching for the generated fingerprints in a local reference database on the media device in response to a search query requested by a client, wherein the local reference database is a subset of a remote reference database;
searching in a local not identified cache on the media device if the generated fingerprints are not found in the local reference database; and
responding to the client with metadata regarding the media content upon finding the generated fingerprints in the local not identified cache.
20. The computer readable non-transitory medium of claim 19 further comprising:
searching the remote reference database if the generated fingerprints are not found in the local not identified cache; and
storing the media content and the generated fingerprints in the local not identified cache upon not finding the generated fingerprints in the remote reference database, wherein subsequent searches for the media content are found in the local not identified cache without searching the remote reference database.
21. The computer readable non-transitory medium of claim 19 further comprising:
searching the remote reference database if the generated fingerprints are not found in the local not identified cache; and
storing the media content and the generated fingerprints in the local reference database upon finding the generated fingerprints in the remote reference database, wherein subsequent searches for the media content are found in the local reference database without searching the remote reference database.