IP Library Granted Patent US 12,579,152
Granted Patent B2
US 12,579,152 · App. 18/530,543 · Granted Mar 17, 2026

Contextual search on multimedia content

Inventors: Gökhan Hasan Bakir (Zurich, CH); Károly Csalogány (Zurich, CH); Behshad Behzadi (Zurich, CH)
Assignee: GOOGLE LLC
G06F16/24575G06F16/43G06F16/951G06F16/9538
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,579,152
App. No.
18/530,543
Granted
Mar 17, 2026
Kind
B2
Abstract

Techniques for contextual search on multimedia content are provided. An example method includes extracting entities associated with multimedia content, wherein the entities include values characterizing one or more objects represented in the multimedia content, generating one or more query rewrite candidates based on the extracted entities and one or more terms in a query related to the multimedia content, providing the one or more query rewrite candidates to a search engine, scoring the one or more query rewrite candidates, ranking the scored one or more query rewrite candidates based on their respective scores, rewriting the query related to the multimedia content based on a particular ranked query rewrite candidate and providing for display, responsive to the query related to the multimedia content, a result set from the search engine based on the rewritten query.

Claims (72)

1 . A method implemented by one or more processors, the method comprising:

identifying a plurality of multimedia content items;

extracting, for a given multimedia content item of the plurality of multimedia content items, one or more corresponding entities, wherein each of the one or more corresponding entities characterize one or more corresponding objects represented in the given multimedia content item, and wherein extracting the one or more corresponding entities for the given multimedia content item further comprises:

extracting one or more corresponding timestamps or time ranges, for each of the one or more corresponding entities, when which each of the one or more corresponding entities appear in the multimedia content, and

storing, in an index database and in association with each of the one or more corresponding entities, the one or more corresponding timestamps or time ranges when which each of the one or more corresponding entities appear in the multimedia content;

storing, in the index database, the one or more corresponding entities; and

subsequent to storing the one or more corresponding entities in the index database:

receiving input associated with a client computing device of a user, the input being associated with the given multimedia content item, and the input being received while the given multimedia content item is being streamed at the client computing device;

generating, based on the one or more corresponding entities for the given multimedia content item that are stored in the index database and based on the input, a query;

executing, based on the query, a search to obtain a result that is responsive to the query; and

causing the result to be provided for presentation to the user responsive to receiving the input and via the client computing device, the result being provided for presentation to the user without interrupting consumption of the given multimedia content item that is being streamed at the client computing device.

2 . The method of claim 1 , wherein extracting the one or more corresponding entities for the given multimedia content item further comprises:

extracting the one or more corresponding entities from metadata associated with the given multimedia content item.

3 . The method of claim 1 , wherein extracting the one or more corresponding entities for the given multimedia content item further comprises:

extracting the one or more corresponding entities from comments or annotations associated with the given multimedia content item.

4 . The method of claim 1 , further comprising:

prior to storing the one or more corresponding entities in the index database:

assigning a corresponding label to each of the one or more corresponding entities.

5 . The method of claim 1 , further comprising:

identifying, from a query repository, one or more historical queries associated with the given multimedia content item;

generating, based on the one or more corresponding entities for the given multimedia content item that are stored in the index database and based on the one or more historical queries, a plurality of rewrites of the one or more historical queries; and

storing, in the index database and in association with each of the one or more corresponding entities, one or more of the plurality of rewrites of the one or more historical queries.

6 . The method of claim 5 , further comprising:

ranking the plurality of rewrites of the one or more historical queries,

wherein storing one or more of the plurality of rewrites of the one or more historical queries comprises:

storing a highest ranking one of the plurality of rewrites of the one or more historical queries in the index database and in association with each of the one or more corresponding entities.

7 . The method of claim 6 , wherein ranking the plurality of rewrites of the one or more historical queries is based on corresponding results obtained responsive to submitting the plurality of rewrites of the one or more historical queries to a search engine.

8 . The method of claim 1 , wherein the one or more corresponding entities characterizing the at least one object represented in the given multimedia content item is one of: a name of a person represented in the given multimedia content item, a name of a person associated with the given multimedia content item, or a name of an object represented in the given multimedia content item.

9 . A system comprising:

memory comprising instructions; and

at least one processor configured to execute the instructions to:

identify a plurality of multimedia content items;

extract, for a given multimedia content item of the plurality of multimedia content items, one or more corresponding entities, wherein each of the one or more corresponding entities characterize one or more corresponding objects represented in the given multimedia content item, wherein, in extracting the one or more corresponding entities for the given multimedia content item, the at least one processor is further configured to execute the instructions to:

extract one or more corresponding timestamps or time ranges, for each of the one or more corresponding entities, when which each of the one or more corresponding entities appear in the multimedia content, and

store, in an index database and in association with each of the one or more corresponding entities, the one or more corresponding timestamps or time ranges when which each of the one or more corresponding entities appear in the multimedia content;

store, in the index database, the one or more corresponding entities; and

subsequent to storing the one or more corresponding entities in the index database:

receive input associated with a client computing device of a user, the input being associated with the given multimedia content item, and the input being received while the given multimedia content item is being streamed at the client computing device;

generate, based on the one or more corresponding entities for the given multimedia content item that are stored in the index database and based on the input, a query;

execute, based on the query, a search to obtain a result that is responsive to the query; and

cause the result to be provided for presentation to the user responsive to receiving the input and via the client computing device, the result being provided for presentation to the user without interrupting consumption of the given multimedia content item that is being streamed at the client computing device.

10 . The system of claim 9 , wherein, in extracting the one or more corresponding entities for the given multimedia content item, the at least one processor is further configured to execute the instructions to:

extract the one or more corresponding entities from metadata associated with the given multimedia content item.

11 . The system of claim 9 , wherein, in extracting the one or more corresponding entities for the given multimedia content item, the at least one processor is further configured to execute the instructions to:

extract the one or more corresponding entities from comments or annotations associated with the given multimedia content item.

12 . The system of claim 9 , wherein the at least one processor is further configured to execute the instructions to:

prior to storing the one or more corresponding entities in the index database:

assign a corresponding label to each of the one or more corresponding entities.

13 . The system of claim 9 , wherein the at least one processor is further configured to execute the instructions to:

identify, from a query repository, one or more historical queries associated with the given multimedia content item;

generate, based on the one or more corresponding entities for the given multimedia content item that are stored in the index database and based on the one or more historical queries, a plurality of rewrites of the one or more historical queries; and

store, in the index database and in association with each of the one or more corresponding entities, one or more of the plurality of rewrites of the one or more historical queries.

14 . The system of claim 13 , wherein the at least one processor is further configured to execute the instructions to:

rank the plurality of rewrites of the one or more historical queries,

wherein, in storing one or more of the plurality of rewrites of the one or more historical queries, the at least one processor is configured to execute the instructions to:

store a highest ranking one of the plurality of rewrites of the one or more historical queries in the index database and in association with each of the one or more corresponding entities.

15 . The system of claim 14 , wherein ranking the plurality of rewrites of the one or more historical queries is based on corresponding results obtained responsive to submitting the plurality of rewrites of the one or more historical queries to a search engine.

16 . The system of claim 9 , wherein the one or more corresponding entities characterizing the at least one object represented in the given multimedia content item is one of: a name of a person represented in the given multimedia content item, a name of a person associated with the given multimedia content item, or a name of an object represented in the given multimedia content item.

17 . The system of claim 9 , wherein, in causing the result to be provided for presentation to the user responsive to receiving the input and via the client computing device, the at least one processor is configured to execute the instructions to:

transmit data to the client computing device that causes the client computing device to provide the result for presentation to the user responsive to receiving the input; or

cause the client computing device to provide the result for presentation to the user responsive to receiving the input.

18 . A non-transitory machine-readable medium comprising instructions stored therein, which when executed by a processor, causes the processor to perform operations, the operations comprising:

identifying a plurality of multimedia content items;

extracting, for a given multimedia content item of the plurality of multimedia content items, one or more corresponding entities, wherein each of the one or more corresponding entities characterize one or more corresponding objects represented in the given multimedia content item, and wherein extracting the one or more corresponding entities for the given multimedia content item further comprises:

extracting one or more corresponding timestamps or time ranges, for each of the one or more corresponding entities, when which each of the one or more corresponding entities appear in the multimedia content, and

storing, in an index database and in association with each of the one or more corresponding entities, the one or more corresponding timestamps or time ranges when which each of the one or more corresponding entities appear in the multimedia content;

storing, in the index database, the one or more corresponding entities; and

subsequent to storing the one or more corresponding entities in the index database:

receiving input associated with a client computing device of a user, the input being associated with the given multimedia content item, and the input being received while the given multimedia content item is being streamed at the client computing device;

generating, based on the one or more corresponding entities for the given multimedia content item that are stored in the index database and based on the input, a query;

executing, based on the query, a search to obtain a result that is responsive to the query; and

causing the result to be provided for presentation to the user responsive to receiving the input and via the client computing device, the result being provided for presentation to the user without interrupting consumption of the given multimedia content item that is being streamed at the client computing device.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 6, 2023
From: BAKIR, GÖKHAN HASAN; CSALOGÁNY, KÁROLY; BEHZADI, BEHSHAD
To: GOOGLE INC.
Reel/Frame 065794/0233 →
CHANGE OF NAME Recorded Dec 6, 2023
From: GOOGLE INC.
To: GOOGLE LLC
Reel/Frame 065794/0236 →
Continuity (4)
Continuation 17531332 · Nov 19, 2021
Continuation 15815349 · Nov 16, 2017
Continuation 14312630 · Jun 23, 2014
Related Publication 20240104104A1 · Mar 28, 2024
References Cited (114)
US 5548758A · Pirahesh et al. · 1996 [cited by applicant]
US 5794189A · Gould · 1998 [cited by applicant]
US 5819286A · Yang et al. · 1998 [cited by applicant]
US 6006241A · Purnaveja et al. · 1999 [cited by applicant]
US 6026389A · Nakajima et al. · 2000 [cited by applicant]
US 6064959A · Young et al. · 2000 [cited by applicant]
US 6081801A · Cochrane et al. · 2000 [cited by applicant]
US 6173317B1 · Chaddha et al. · 2001 [cited by applicant]
US 6175829B1 · Li et al. · 2001 [cited by applicant]
US 6182069B1 · Niblack et al. · 2001 [cited by applicant]
US 6272495B1 · Hetherington · 2001 [cited by applicant]
US 6314419B1 · Faisal · 2001 [cited by examiner]
US 6480843B2 · Li · 2002 [cited by examiner]
US 7050110B1 · Lienhart et al. · 2006 [cited by applicant]
US 7334186B2 · Wu · 2008 [cited by applicant]
US 7650623B2 · Hudgeons et al. · 2010 [cited by applicant]
US 7827254B1 · Nevill-Manning et al. · 2010 [cited by applicant]
US 7840590B2 · Tam et al. · 2010 [cited by applicant]
US 7856429B2 · Herberger et al. · 2010 [cited by applicant]
US 8081956B2 · Aaltonen et al. · 2011 [cited by applicant]
US 8122103B2 · Grimault et al. · 2012 [cited by applicant]
US 8135619B2 · Bem · 2012 [cited by applicant]
US 8433611B2 · Lax et al. · 2013 [cited by applicant]
US 8588525B1 · Yagnik · 2013 [cited by examiner]
US 8635287B1 · Shih · 2014 [cited by examiner]
US 8650210B1 · Cheng et al. · 2014 [cited by applicant]
US 8914394B1 · Dubois et al. · 2014 [cited by applicant]
US 8959071B2 · Gunatilake · 2015 [cited by applicant]
US 8972875B2 · Gossweiler, III et al. · 2015 [cited by applicant]
US 9547690B2 · Nowak-Przygodzki et al. · 2017 [cited by applicant]
US 9852188B2 · Bakir et al. · 2017 [cited by applicant]
US 11204927B2 · Bakir et al. · 2021 [cited by applicant]
US 20020131641A1 · Luo · 2002 [cited by examiner]
US 20020161747A1 · Li · 2002 [cited by examiner]
US 20020188602A1 · Stubler et al. · 2002 [cited by applicant]
US 20030233375A1 · Sagar · 2003 [cited by applicant]
US 20040098376A1 · Li et al. · 2004 [cited by applicant]
US 20040161747A1 · Morahan · 2004 [cited by examiner]
US 20040205461A1 · Kaufman · 2004 [cited by examiner]
US 20040210443A1 · Kuhn · 2004 [cited by examiner]
US 20070061317A1 · Ramer et al. · 2007 [cited by applicant]
US 20080147614A1 · Tam et al. · 2008 [cited by applicant]
US 20090113475A1 · Li · 2009 [cited by applicant]
US 20090144312A1 · Patel · 2009 [cited by applicant]
US 20100076840A1 · Bem · 2010 [cited by examiner]
US 20100121864A1 · Stuhec et al. · 2010 [cited by applicant]
US 20100145971A1 · Cheng et al. · 2010 [cited by applicant]
US 20100185611A1 · Liao et al. · 2010 [cited by applicant]
US 20100205203A1 · Anderson et al. · 2010 [cited by applicant]
US 20100228776A1 · Melkote et al. · 2010 [cited by applicant]
US 20110295847A1 · Cucerzan · 2011 [cited by applicant]
US 20120084312A1 · Jenson · 2012 [cited by applicant]
US 20120131031A1 · Xie et al. · 2012 [cited by applicant]
US 20120131641A1 · Kavantzas et al. · 2012 [cited by applicant]
US 20120226696A1 · Thambiratnam et al. · 2012 [cited by applicant]
US 20120233147A1 · Solheim et al. · 2012 [cited by applicant]
US 20120323897A1 · Daher · 2012 [cited by examiner]
US 20130086105A1 · Hammontree · 2013 [cited by examiner]
US 20130311504A1 · Xie · 2013 [cited by examiner]
US 20140136640A1 · Shih · 2014 [cited by examiner]
US 20140330804A1 · Bao et al. · 2014 [cited by applicant]
US 20150206070A1 · Kulkarni · 2015 [cited by examiner]
US 20150227840A1 · Codella et al. · 2015 [cited by applicant]
US 20150379079A1 · Kota · 2015 [cited by applicant]
US 20160125028A1 · Silvestri et al. · 2016 [cited by applicant]
US 20190144312A1 · King · 2019 [cited by applicant]
US 20220075787A1 · Bakir et al. · 2022 [cited by applicant]
CN 101021857 · 2007 [cited by applicant]
CN 101334796 · 2008 [cited by applicant]
CN 101842787 · 2010 [cited by applicant]
CN 102402593 · 2012 [cited by applicant]
CN 102521321 · 2012 [cited by applicant]
CN 102999560 · 2013 [cited by applicant]
CN 106663104 · 2017 [cited by applicant]
JP H0368074 · 1991 [cited by applicant]
JP H07302267 · 1995 [cited by applicant]
JP 2009301498 · 2009 [cited by applicant]
RU 2378680 · 2010 [cited by applicant]
RU 2010121864 · 2011 [cited by applicant]
WO 2007081682 · 2007 [cited by applicant]
WO 2012158542 · 2012 [cited by applicant]
WO 2015199842 · 2015 [cited by applicant]
Porkaew et al.: Query reformulation for contant based multimedia retrieval in MARS, IEEE 1999. [cited by examiner]
Porkaew et al., Query reformation for constant based multimedia retrieval in MARS, IEEE 1999. [cited by examiner]
Porkaew et al., “Query Reformulation for Content Based Multimedia Retrieval on MARS”, IEEE 1999. [cited by examiner]
China National Intellectual Property Administration; Notification of First Office Action issued in Application No. 202010483818.9; 28 pages; dated Jun. 17, 2024. [cited by applicant]
Porkaew et al., “Query Reformulation for Content Based Multimedia Retrieval on MARS,” University of Illinois, 1998. 18 pages. 1998. [cited by applicant]
Porkaew et al., “Query Reformulation for Content Based Multimedia Retrieval in Mars”, IEEE International Conference on Multimedia Computing and Systems, Florence, Italy, Jun. 7-11, 1999, pp. 747-751. [cited by applicant]
Tan, et al., “A Framework for Measuring Video Similarity and Its Application to Video Query by Example,” 1999 International Conference on ( vol. 2) Image Processing, 1999, pp. 106-110 vol. 2 1999. [cited by applicant]
Ahanger, et al., “Video Query Formulation,” Multimedia Communications Laboratory Department of Electrical, Computer and Systems Engineering, Boston University, Jan. 9, 1995, pp. 1-21. 1995. [cited by applicant]
Flickner, et al., “Query by Image and Video Content: The QBIC System,” IBM Almaden Research Center, San Jose, CA, Sep. 1995, pp. 23-32. 1995. [cited by applicant]
Liu, et al. “A New Approach to Retrieve Video by Example Video Clip,” Proceeding of the 7th ACM International Multimedia Conference (Multimedia 99), Orlando, Flordia, 1999, pp. 1-4. [cited by applicant]
International Search Report and Written Opinion of PCT Ser. No. PCT/US2015/031362; 10 pages Aug. 6, 2015. [cited by applicant]
Japanese Patent Office; Office Action issued in Application No. 2016-574968 dated Feb. 9, 2018. [cited by applicant]
Japanese Patent Office; Notice of Allowance issued in Application No. 2016-574968 dated Sep. 10, 2018 Sep. 10, 2018. [cited by applicant]
Intellectual Property of India; Hearing Notice issued in Application No. 201647044405; 3 pages; dated Aug. 17, 2022. [cited by applicant]
Natsev et al., “Semantic Concept-Based Query Expansion and Re-ranking for Multimedia Retrieval,” ACM 2007. [cited by applicant]
Patent Office of the Russian Federation; Office Action and Search Report with English translation issued in Application No. 2016150419; 19 pages; dated Nov. 21, 2018. [cited by applicant]
China National Intellectual Property Administration; Office Action issued in Application No. 20158003459.3 dated Mar. 25, 2019. [cited by applicant]
Patent Office of the Russian Federation; Decision to Grant issued in Application No. 2016150419 dated Apr. 16, 2019. [cited by applicant]
European Patent Office; Examination Report issued in Application No. 15727173.5 dated Aug. 8, 2019. [cited by applicant]
China National Intellectual Property Administration; Office Action issued in Application No. 201580034359.3 dated Sep. 30, 2019. [cited by applicant]
Japanese Patent Office; Office Action issued in Application No. 2018-191650 dated Jan. 6, 2020. [cited by applicant]
Patent Office of the Russian Federation; Office Action issued in Application No. 2019119697, 9 pages, dated Jan. 30, 2020. [cited by applicant]
European Patent Office; Summons to Attend Oral Proceedings issued in Application No. 15727173.5 dated Feb. 19, 2020. [cited by applicant]
China National Intellectual Property Administration; Notice of Grant issued in Application No. 201580034359.3; 4 pages; dated Mar. 17, 2020. [cited by applicant]
Kraft, Reiner et al.; Searching wiht Context; Yahoo! Inc.; IEEE; pp. 477-486; dated 2006. [cited by applicant]
Patent Office of the Russian Federation; Decision to Grant issued in Application No. 2019119697; 18 pages, dated May 20, 2020. [cited by applicant]
Intellectual Property India; Office Action issue in Applicaion No. 201647044405; 8 pages; dated Apr. 20, 2020. [cited by applicant]
European Patent Office: Oral Proceedings issued for Application No. 15727173.5 dated Jun. 19, 2020. [cited by applicant]
Japanese Patent Office; Notice of Reasons for Rejection issued in Application No. 2018-191650; 6 pages; dated Aug. 31, 2020. [cited by applicant]
Japanese Patent Office; Notice of Allowance issued in Application No. 2018-191650; 3 pages; dated May 10, 2021. [cited by applicant]
China National Intellectual Property Administration; Notification of Second Office Action issued in Application No. 202010483818.9; 10 pages; dated Nov. 29, 2024. [cited by applicant]
China National Intellectual Property Administration; Notice of Grant issued in Application No. 202010483818.9; 4 pages; dated Mar. 28, 2025. [cited by applicant]