公开(公告)号：US20230362117A1

公开(公告)日：2023-11-09

申请号：US18351894

申请日：2023-07-13

申请人： Snap Inc.

发明人： Jean Luo ,  Celia Nicole Mourkogiannis

IPC分类号： H04L51/046 ,  G06Q30/0601 ,  G06Q20/38 ,  G06F16/735 ,  G06F16/783 ,  H04L51/52

CPC分类号： H04L51/046 ,  G06Q30/0643 ,  G06Q30/0631 ,  G06Q20/386 ,  G06F16/735 ,  G06F16/7837 ,  G06Q20/384 ,  H04L51/52

摘要： In one or more implementations, video content generated using a client application may be provided to users of the client application in association with one or more products offered for purchase via the client application. Video content items presented to a user may be identified based on a level of interest of the user in relation to the respective video content items. A level of interest of the user in relation to a video content item may be based on profile information of the user.

公开(公告)号：US20230325438A1

公开(公告)日：2023-10-12

申请号：US17891486

申请日：2022-08-19

申请人： ELECTRONICS AND TELECOMMUNICATIONS RESERCH INSTITUTE

发明人： Jong Hee KIM ,  Hyung II KIM ,  Jin Young MOON ,  Je Seok HAM

IPC分类号： G06F16/78 ,  G06F40/279 ,  G06F16/735 ,  G06F16/738

CPC分类号： G06F16/7867 ,  G06F40/279 ,  G06F16/735 ,  G06F16/738

摘要： Disclosed herein are a method and apparatus for searching for a video section by using a natural language. The method for searching for a video section includes: extracting keywords from a natural language sentence, when the natural language sentence is input; determining whether or not the extracted keywords are included in predefined context information; and deriving and providing a final search result. In addition, when the extracted keywords are included in the predefined context information, a search result is derived by performing a first method, and when the extracted keywords are not included in the predefined context information, a search result is derived by performing a second method.

公开(公告)号：US11769207B2

公开(公告)日：2023-09-26

申请号：US17347629

申请日：2021-06-15

申请人： Theator Inc.

发明人： Tamir Wolf ,  Dotan Asselmann

IPC分类号： G06K9/00 ,  G06Q40/08 ,  A61B90/00 ,  A61B34/00 ,  A61B34/37 ,  G16H20/40 ,  G16H50/70 ,  G16H50/20 ,  G16H30/40 ,  G16H70/20 ,  G16H15/00 ,  A61B34/30 ,  G16H40/63 ,  A61B17/00 ,  G16H50/30 ,  G06F16/71 ,  A61B34/10 ,  G16H40/20 ,  G16H30/20 ,  A61B5/02 ,  A61B5/103 ,  A61B5/00 ,  G06Q10/0631 ,  G06V20/20 ,  G06V20/40 ,  G06V20/52 ,  A61B90/92 ,  G06F16/735 ,  G11B27/10 ,  A61B1/00 ,  G06V40/10 ,  A61B1/04

CPC分类号： G06Q40/08 ,  A61B1/000094 ,  A61B1/000096 ,  A61B5/02042 ,  A61B5/1032 ,  A61B5/7267 ,  A61B5/7275 ,  A61B5/746 ,  A61B17/00 ,  A61B34/10 ,  A61B34/25 ,  A61B34/30 ,  A61B34/37 ,  A61B34/70 ,  A61B90/06 ,  A61B90/361 ,  A61B90/37 ,  A61B90/92 ,  G06F16/71 ,  G06F16/735 ,  G06Q10/06312 ,  G06V20/20 ,  G06V20/40 ,  G06V20/41 ,  G06V20/49 ,  G06V20/52 ,  G11B27/10 ,  G16H15/00 ,  G16H20/40 ,  G16H30/20 ,  G16H30/40 ,  G16H40/20 ,  G16H40/63 ,  G16H50/20 ,  G16H50/30 ,  G16H50/70 ,  G16H70/20 ,  A61B1/04 ,  A61B2017/0011 ,  A61B2017/00119 ,  A61B2034/104 ,  A61B2034/252 ,  A61B2034/254 ,  A61B2034/256 ,  A61B2034/302 ,  A61B2090/065 ,  A61B2090/0807 ,  A61B2505/05 ,  G06V20/44 ,  G06V40/10 ,  G06V2201/034

摘要： Systems and methods for automatically populating a post-operative report of a surgical procedure are disclosed. A system may include at least one processor configured to implement a method including receiving an identifier of a patient, an identifier of a healthcare provider, and surgical footage of a surgical procedure performed on the patient. The method may include analyzing frames of the surgical footage to identify phases of the surgical procedure based on interactions between medical instruments and biological structures and, based on the interactions, associate a name with each phase. The method may include determining a beginning of each phase and associating a time marker with the beginning of each phase. The method may include populating a post-operative report with the patient identifier, the names of the phases, and time markers associated with the phases in a manner that enables the health care provider to alter the post-operative report.

公开(公告)号：US11763564B1

公开(公告)日：2023-09-19

申请号：US17216147

申请日：2021-03-29

申请人： Amazon Technologies, Inc.

发明人： Najmeh Sadoughi Nourabadi ,  Kewen Chen ,  Tu Anh Ho ,  Christina Botkins ,  Dongqing Zhang ,  Muhammad Raffay Hamid

IPC分类号： G06V20/40 ,  G06F16/71 ,  G06F16/75 ,  G06F16/735 ,  G06N20/00

CPC分类号： G06V20/46 ,  G06F16/71 ,  G06F16/735 ,  G06F16/75 ,  G06N20/00 ,  G06V20/41

摘要： Systems and methods are provided herein for generating optimized video segments. A derivative video segment (e.g., a scene) can be identified from derivative video content (e.g., a movie trailer). The segment may be used a query to search video content (e.g., the movie) for the segment. Once found, an optimized video segment may be generated from the video content. The optimized video segment may have a different start time and/or end time than those corresponding to the original segment. Once optimized, the video segment may be presented to a user or stored for subsequent content recommendations.

公开(公告)号：US11736742B2

公开(公告)日：2023-08-22

申请号：US16870460

申请日：2020-05-08

申请人： Roku, Inc.

发明人： Andrew Laurence Ashbacher ,  Steven Michael Cormie ,  Joel Michael Curl ,  John L. Haynes, III

IPC分类号： H04N21/234 ,  G06F16/735 ,  G06F16/783 ,  H04N21/8358 ,  H04N21/4402 ,  H04N21/2389 ,  G06F16/738 ,  H04N21/435 ,  H04N21/442

CPC分类号： H04N21/23418 ,  G06F16/735 ,  G06F16/738 ,  G06F16/783 ,  H04N21/23892 ,  H04N21/435 ,  H04N21/442 ,  H04N21/4402 ,  H04N21/8358

摘要： In one aspect, a method is for use in connection with a content-modification system that includes a content-distribution system and a content-presentation device. The method includes (i) identifying an upcoming content modification opportunity on an identified channel, wherein the identifying is based on detecting a match between first reference fingerprint data representing an initial portion of a modifiable content-segment and query fingerprint data representing content transmitted by a content-distribution system to a content-presentation device, wherein the first reference fingerprint data was generated before the query fingerprint data was generated; and (ii) responsive to identifying the upcoming content modification opportunity, transmitting to the content-presentation device, second reference fingerprint data representing more than the initial portion of the modifiable content-segment to facilitate the content-presentation device to, at a later time, abort performance of a partially performed content-modification operation related to the identified content modification opportunity.

公开(公告)号：US20230237093A1

公开(公告)日：2023-07-27

申请号：US17649091

申请日：2022-01-27

申请人： ADOBE INC.

发明人： Yifan Li ,  Trung Huu Bui ,  Timothy Jeewun Ganter ,  David Fox

IPC分类号： G06F16/735 ,  G06N3/08

CPC分类号： G06F16/735 ,  G06N3/08

摘要： Systems and methods for item recommendation are described. Embodiments of the present disclosure receive input indicating a relationship between a user and a first content item; generate a knowledge graph based on the input, wherein the knowledge graph comprises relationship information between the user and a plurality of content items; generate a first feature embedding representing the user and a second feature embedding representing a second content item of the plurality of content items based on the knowledge graph, wherein the second feature embedding is generated using a first modality for a query vector of an attention mechanism and a second modality for a key vector and a value vector of the attention mechanism; compare the first feature embedding to the second feature embedding to obtain a similarity score; and recommend the second content item for the user based on the similarity score.

公开(公告)号：US20230229698A1

公开(公告)日：2023-07-20

申请号：US18123542

申请日：2023-03-20

申请人： Google LLC

发明人： Luca Chiarandini ,  Krystof Hoder

IPC分类号： G06F16/735 ,  G06F16/738 ,  G06F16/75

CPC分类号： G06F16/735 ,  G06F16/738 ,  G06F16/75

摘要： Methods, systems, and media for demoting search results based on search query risk are provided. In some embodiments, a method for demoting search results includes: receiving a search query for a video content item; generating a plurality of search results in response to the search query; calculating a set of result goodness values; calculating a query goodness value associated with the search query based on the set of result goodness values; identifying a threshold goodness value based on the query goodness value; in response to determining that a first result goodness value of the set of result goodness values is less than the threshold goodness value, demoting a first search result of the plurality of search results, wherein the first result goodness value is associated with the first search result; and causing at least a portion of the plurality of search results to be presented based on the demotion.

公开(公告)号：US20230222235A1

公开(公告)日：2023-07-13

申请号：US17572968

申请日：2022-01-11

申请人： Kyocera Document Solutions, Inc.

发明人： Edgar Nielzen Alconis QUIBAL ,  Vicente Cheng GUTIERREZ

IPC分类号： G06F21/62 ,  G11B27/02 ,  G06F16/735

CPC分类号： G06F21/6218 ,  G06F16/735 ,  G11B27/02

摘要： A system and method for redacting private or sensitive information in a video file before publishing the video file to the public are disclosed. The system and method do not redact an original video file but a duplicated video file, and only redacted duplicated video file will be displayed to a public user. A system and method for retrieving original data corresponding to the redacted data are also disclosed. Metadata is embedded in redacted data of the redacted duplicated video file, which when activated, will direct to a location where the original data is stored.

公开(公告)号：US11681749B2

公开(公告)日：2023-06-20

申请号：US17528871

申请日：2021-11-17

申请人： Snap Inc.

发明人： Nathan Kenneth Boyd ,  Maryam Daneshi ,  Sivakumar Loganathan ,  Satya Kanduri ,  Senthil Karthikeyan Sundaram ,  Evan Shaw

IPC分类号： G06F16/735 ,  G06F16/2457 ,  G06F16/78 ,  G06F16/71 ,  G06V20/40

CPC分类号： G06F16/735 ,  G06F16/24578 ,  G06F16/71 ,  G06F16/7867 ,  G06V20/41 ,  G06V2201/10

摘要： Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing at least one program and a method for ranking videos. The program and method include storing, in a database, a plurality of videos; obtaining, for a first of the plurality of videos, first data representing a first interaction performed by a creator of the first video in association with the first video; and obtaining, for the first video, second data representing a second interaction performed by a consumer of the first video in association with the first video. A score is computed for the first video based on the first and second data and the first video is ranked relative to a remaining subset of the plurality of videos by comparing the computed score for the first video with respective scores of videos in the remaining subset of the plurality of videos.

公开(公告)号：US11675835B2

公开(公告)日：2023-06-13

申请号：US16826101

申请日：2020-03-20

申请人： BITCENTRAL, INC.

发明人： Daniel Goikhman ,  Krishna Arvapally ,  Latha Maduri Doddikadi ,  Akshay Arvapally

IPC分类号： G06F16/78 ,  H04N21/258 ,  H04N21/235 ,  H04N21/2668 ,  H04N21/234 ,  H04N21/466 ,  G06F16/735 ,  G06F16/738 ,  G06F16/951 ,  H04N21/231 ,  H04N21/482 ,  H04N21/81 ,  G06V20/40 ,  H04N21/475 ,  H04N21/4788 ,  H04N21/84 ,  H04N21/61

CPC分类号： G06F16/7867 ,  G06F16/735 ,  G06F16/738 ,  G06F16/951 ,  G06V20/41 ,  H04N21/235 ,  H04N21/23113 ,  H04N21/23418 ,  H04N21/25891 ,  H04N21/2668 ,  H04N21/4668 ,  H04N21/4756 ,  H04N21/4788 ,  H04N21/4828 ,  H04N21/6125 ,  H04N21/8133 ,  H04N21/84 ,  G06V20/44

摘要： A video ingest, search and delivery solution is provided in which videos and their associated metadata are ingested based on ingest criteria. Moments in the metadata are identified by parsing comments in the metadata for comments that reference a time in the video. Moments are recorded with the referenced time and the text of the comment. Videos are rated and may be delivered to users based at least in part on the rating. Moments may be displayed together with the video by displaying the text of moments at the associated time in the video as the video is being delivered. The text of moment records may be searched based on search parameters received form a user to identify videos to deliver to the user.