-
公开(公告)号：US11917332B2
公开(公告)日：2024-02-27
申请号：US16136647
申请日：2018-09-20
Applicant: Comcast Cable Communications, LLC
Inventor： Evelyne Tzoukermann , Hongzhong Zhou , Oliver Jojic , Bageshree Shevade , Ryan M. Farrell
IPC: H04N21/234 , H04N21/44 , H04N7/173 , H04N21/233 , H04N21/433 , H04N21/439 , H04N21/472 , H04N21/81 , H04N21/84 , H04N21/845 , H04N21/218
CPC classification number: H04N7/17318 , H04N21/233 , H04N21/23418 , H04N21/23424 , H04N21/4334 , H04N21/4394 , H04N21/44008 , H04N21/44016 , H04N21/47202 , H04N21/812 , H04N21/84 , H04N21/8456 , H04N21/21815
Abstract: Content streams may be segmented to provide automatic extraction and storage of content items without intervening commercials or other unrelated content. These content items may then be stored in a database and made accessible to subscribers through, for example, an on-demand service. Automatic segmentation may include the identification of program boundaries, segmentation of a content stream based on the boundaries and the subsequent classification of the segments into content types. For example, audio and video duplication detection may be used to identify commercials since commercials tend to repeat frequently over a relatively short amount of time. A system may further identify an end of program indicator in a video stream to determine when a program ends. Accordingly, if a program ends after a scheduled end time, a recording device (e.g., the program is being recorded) may automatically extend the recording time to capture the entire program.
-
公开(公告)号：US20170091556A1
公开(公告)日：2017-03-30
申请号：US15166747
申请日：2016-05-27
Applicant: Comcast Cable Communications, LLC
Inventor： Jan Neumann , Evelyne Tzoukermann , Amit Bagga , Oliver Jojic , Bageshree Shevade , David F. Houghton , Corey Farrell
CPC classification number: G06K9/00718 , G06F17/275 , G06F17/30781 , G06K9/00 , G06K9/00221 , G06K9/00228 , G06K9/00744 , G06K9/00765 , G06K9/34 , G06K9/48 , G06K9/481 , G10L15/22 , G10L25/57
Abstract: The disclosure relates to recognizing data such as items or entities in content. In some aspects, content may be received and feature information, such as face recognition data and voice recognition data may be generated. Scene segmentation may also be performed on the content, grouping the various shots of the video content into one or more shot collections, such as scenes. For example, a decision lattice representative of possible scene segmentations may be determined and the most probable path through the decision lattice may be selected as the scene segmentation. Upon generating the feature information and performing the scene segmentation, one or more items or entities that are present in the scene may be identified.
-
公开(公告)号：US20190260969A1
公开(公告)日：2019-08-22
申请号：US16136647
申请日：2018-09-20
Applicant: Comcast Cable Communications, LLC
Inventor： Evelyne Tzoukermann , Hongzhong Zhou , Oliver Jojic , Bageshree Shevade , Ryan M. Farrell
IPC: H04N7/173 , H04N21/44 , H04N21/234 , H04N21/845 , H04N21/84 , H04N21/81 , H04N21/472 , H04N21/233 , H04N21/439 , H04N21/433
Abstract: Content streams may be segmented to provide automatic extraction and storage of content items without intervening commercials or other unrelated content. These content items may then be stored in a database and made accessible to subscribers through, for example, an on-demand service. Automatic segmentation may include the identification of program boundaries, segmentation of a content stream based on the boundaries and the subsequent classification of the segments into content types. For example, audio and video duplication detection may be used to identify commercials since commercials tend to repeat frequently over a relatively short amount of time. A system may further identify an end of program indicator in a video stream to determine when a program ends. Accordingly, if a program ends after a scheduled end time, a recording device (e.g., the program is being recorded) may automatically extend the recording time to capture the entire program.
-
公开(公告)号：US09378423B2
公开(公告)日：2016-06-28
申请号：US14475742
申请日：2014-09-03
Applicant: Comcast Cable Communications, LLC
Inventor： Jan Neumann , Evelyne Tzoukermann , Amit Bagga , Oliver Jojic , Bageshree Shevade , David F. Houghton , Corey Farrell
CPC classification number: G06K9/00718 , G06F17/275 , G06F17/30781 , G06K9/00 , G06K9/00221 , G06K9/00228 , G06K9/00744 , G06K9/00765 , G06K9/34 , G06K9/48 , G06K9/481 , G10L15/22 , G10L25/57
Abstract: The disclosure relates to recognizing data such as items or entities in content. In some aspects, content may be received and feature information, such as face recognition data and voice recognition data may be generated. Scene segmentation may also be performed on the content, grouping the various shots of the video content into one or more shot collections, such as scenes. For example, a decision lattice representative of possible scene segmentations may be determined and the most probable path through the decision lattice may be selected as the scene segmentation. Upon generating the feature information and performing the scene segmentation, one or more items or entities that are present in the scene may be identified.
Abstract translation: 本公开涉及识别诸如内容中的项目或实体的数据。 在一些方面，可以接收内容，并且可以生成诸如面部识别数据和语音识别数据的特征信息。 还可以对内容执行场景分割，将视频内容的各种照片分组成一个或多个拍摄集合，诸如场景。 例如，可以确定表示可能的场景分割的决策点，并且可以选择通过决策点的最可能的路径作为场景分割。 在生成特征信息并执行场景分割时，可以识别场景中存在的一个或多个项目或实体。
-
公开(公告)号：US20240205373A1
公开(公告)日：2024-06-20
申请号：US18413857
申请日：2024-01-16
Applicant: Comcast Cable Communications, LLC
Inventor： Evelyne Tzoukermann , Hongzhong Zhou , Oliver Jojic , Bageshree Shevade , Ryan M. Farrell
IPC: H04N7/173 , H04N21/218 , H04N21/233 , H04N21/234 , H04N21/433 , H04N21/439 , H04N21/44 , H04N21/472 , H04N21/81 , H04N21/84 , H04N21/845
CPC classification number: H04N7/17318 , H04N21/233 , H04N21/23418 , H04N21/23424 , H04N21/4334 , H04N21/4394 , H04N21/44008 , H04N21/44016 , H04N21/47202 , H04N21/812 , H04N21/84 , H04N21/8456 , H04N21/21815
Abstract: Content streams may be segmented to provide automatic extraction and storage of content items without intervening commercials or other unrelated content. These content items may then be stored in a database and made accessible to subscribers through, for example, an on-demand service. Automatic segmentation may include the identification of program boundaries, segmentation of a content stream based on the boundaries and the subsequent classification of the segments into content types. For example, audio and video duplication detection may be used to identify commercials since commercials tend to repeat frequently over a relatively short amount of time. A system may further identify an end of program indicator in a video stream to determine when a program ends. Accordingly, if a program ends after a scheduled end time, a recording device (e.g., the program is being recorded) may automatically extend the recording time to capture the entire program.
-
公开(公告)号：US10192116B2
公开(公告)日：2019-01-29
申请号：US15166747
申请日：2016-05-27
Applicant: Comcast Cable Communications, LLC
Inventor： Jan Neumann , Evelyne Tzoukermann , Amit Bagga , Oliver Jojic , Bageshree Shevade , David F. Houghton , Corey Farrell
Abstract: The disclosure relates to recognizing data such as items or entities in content. In some aspects, content may be received and feature information, such as face recognition data and voice recognition data may be generated. Scene segmentation may also be performed on the content, grouping the various shots of the video content into one or more shot collections, such as scenes. For example, a decision lattice representative of possible scene segmentations may be determined and the most probable path through the decision lattice may be selected as the scene segmentation. Upon generating the feature information and performing the scene segmentation, one or more items or entities that are present in the scene may be identified.
-
公开(公告)号：US20150057995A1
公开(公告)日：2015-02-26
申请号：US14475742
申请日：2014-09-03
Applicant: Comcast Cable Communications, LLC
Inventor： Jan Neumann , Evelyne Tzoukermann , Amit Bagga , Oliver Jojic , Bageshree Shevade , David F. Houghton , Corey Farrell
CPC classification number: G06K9/00718 , G06F17/275 , G06F17/30781 , G06K9/00 , G06K9/00221 , G06K9/00228 , G06K9/00744 , G06K9/00765 , G06K9/34 , G06K9/48 , G06K9/481 , G10L15/22 , G10L25/57
Abstract: The disclosure relates to recognizing data such as items or entities in content. In some aspects, content may be received and feature information, such as face recognition data and voice recognition data may be generated. Scene segmentation may also be performed on the content, grouping the various shots of the video content into one or more shot collections, such as scenes. For example, a decision lattice representative of possible scene segmentations may be determined and the most probable path through the decision lattice may be selected as the scene segmentation. Upon generating the feature information and performing the scene segmentation, one or more items or entities that are present in the scene may be identified.
Abstract translation: 本公开涉及识别诸如内容中的项目或实体的数据。 在一些方面，可以接收内容，并且可以生成诸如面部识别数据和语音识别数据的特征信息。 还可以对内容执行场景分割，将视频内容的各种照片分组成一个或多个拍摄集合，诸如场景。 例如，可以确定表示可能的场景分割的决策点，并且可以选择通过决策点的最可能的路径作为场景分割。 在生成特征信息并执行场景分割时，可以识别场景中存在的一个或多个项目或实体。
