-
公开(公告)号:US20210144442A1
公开(公告)日:2021-05-13
申请号:US17095486
申请日:2020-11-11
Applicant: Netflix, Inc.
Inventor: Dong Liu , Lezi Wang , Rohit Puri
IPC: H04N21/466 , H04N21/845 , G06N20/00 , H04N21/44 , G06K9/00
Abstract: The disclosed computer-implemented method may include accessing media segments that correspond to respective media items. At least one of the media segments may be divided into discrete video shots. The method may also include matching the discrete video shots in the media segments to corresponding video shots in the corresponding media items according to various matching factors. The method may further include generating a relative similarity score between the matched video shots in the media segments and the corresponding video shots in the media items, and training a machine learning model to automatically identify video shots in the media items according to the generated relative similarity score between matched video shots. Various other methods, systems, and computer-readable media are also disclosed.
-
公开(公告)号:US20200151546A1
公开(公告)日:2020-05-14
申请号:US16677161
申请日:2019-11-07
Applicant: Netflix, Inc.
Inventor: Dong Liu , Nagendra Kamath , Rohit Puri , Subhabrata Bhattacharya
Abstract: The disclosed computer-implemented method may include generating a three-dimensional (3D) feature map for a digital image using a fully convolutional network (FCN). The 3D feature map may be configured to identify features of the digital image and identify an image region for each identified feature. The method may also include generating a region composition graph that includes the identified features and image regions. The region composition graph may be configured to model mutual dependencies between features of the 3D feature map. The method may further include performing a graph convolution on the region composition graph to determine a feature aesthetic value for each node according to the weightings in the node's weighted connecting segments, and calculating a weighted average for each node's feature aesthetic value to provide a combined level of aesthetic appeal for the digital image. Various other methods, systems, and computer-readable media are also disclosed.
-
公开(公告)号:US20230290383A1
公开(公告)日:2023-09-14
申请号:US18320877
申请日:2023-05-19
Applicant: Netflix, Inc.
Inventor: Dong Liu , Lezi Wang , Rohit Puri
IPC: G11B27/28 , H04N21/466 , H04N21/845 , G06N20/00 , H04N21/44 , G06V20/40 , G06V10/82 , G06V10/44
CPC classification number: G11B27/28 , H04N21/4668 , H04N21/845 , G06N20/00 , H04N21/4667 , H04N21/44008 , G06V20/41 , G06V20/46 , G06V10/82 , G06V10/454 , G06V20/47 , G06V20/48 , G06V20/49
Abstract: The disclosed computer-implemented method may include accessing media segments that correspond to respective media items. At least one of the media segments may be divided into discrete video shots. The method may also include matching the discrete video shots in the media segments to corresponding video shots in the corresponding media items according to various matching factors. The method may further include generating a relative similarity score between the matched video shots in the media segments and the corresponding video shots in the media items, and training a machine learning model to automatically identify video shots in the media items according to the generated relative similarity score between matched video shots. Various other methods, systems, and computer-readable media are also disclosed.
-
公开(公告)号:US11694726B2
公开(公告)日:2023-07-04
申请号:US17725536
申请日:2022-04-20
Applicant: Netflix, Inc.
Inventor: Dong Liu , Lezi Wang , Rohit Puri
IPC: H04N21/466 , G11B27/28 , H04N21/845 , G06N20/00 , H04N21/44 , G06V20/40 , G06V10/82 , G06V10/44
CPC classification number: G11B27/28 , G06N20/00 , G06V10/454 , G06V10/82 , G06V20/41 , G06V20/46 , G06V20/47 , G06V20/48 , G06V20/49 , H04N21/44008 , H04N21/4667 , H04N21/4668 , H04N21/845
Abstract: The disclosed computer-implemented method may include accessing media segments that correspond to respective media items. At least one of the media segments may be divided into discrete video shots. The method may also include matching the discrete video shots in the media segments to corresponding video shots in the corresponding media items according to various matching factors. The method may further include generating a relative similarity score between the matched video shots in the media segments and the corresponding video shots in the media items, and training a machine learning model to automatically identify video shots in the media items according to the generated relative similarity score between matched video shots. Various other methods, systems, and computer-readable media are also disclosed.
-
公开(公告)号:US11551060B2
公开(公告)日:2023-01-10
申请号:US16677161
申请日:2019-11-07
Applicant: Netflix, Inc.
Inventor: Dong Liu , Nagendra Kamath , Rohit Puri , Subhabrata Bhattacharya
Abstract: The disclosed computer-implemented method may include generating a three-dimensional (3D) feature map for a digital image using a fully convolutional network (FCN). The 3D feature map may be configured to identify features of the digital image and identify an image region for each identified feature. The method may also include generating a region composition graph that includes the identified features and image regions. The region composition graph may be configured to model mutual dependencies between features of the 3D feature map. The method may further include performing a graph convolution on the region composition graph to determine a feature aesthetic value for each node according to the weightings in the node's weighted connecting segments, and calculating a weighted average for each node's feature aesthetic value to provide a combined level of aesthetic appeal for the digital image. Various other methods, systems, and computer-readable media are also disclosed.
-
公开(公告)号:US20220248096A1
公开(公告)日:2022-08-04
申请号:US17725536
申请日:2022-04-20
Applicant: Netflix, Inc.
Inventor: Dong Liu , Lezi Wang , Rohit Puri
IPC: H04N21/466 , H04N21/845 , G06N20/00 , H04N21/44 , G06V20/40
Abstract: The disclosed computer-implemented method may include accessing media segments that correspond to respective media items. At least one of the media segments may be divided into discrete video shots. The method may also include matching the discrete video shots in the media segments to corresponding video shots in the corresponding media items according to various matching factors. The method may further include generating a relative similarity score between the matched video shots in the media segments and the corresponding video shots in the media items, and training a machine learning model to automatically identify video shots in the media items according to the generated relative similarity score between matched video shots. Various other methods, systems, and computer-readable media are also disclosed.
-
公开(公告)号:US11350169B2
公开(公告)日:2022-05-31
申请号:US17095486
申请日:2020-11-11
Applicant: Netflix, Inc.
Inventor: Dong Liu , Lezi Wang , Rohit Puri
IPC: H04N21/466 , H04N21/845 , G06N20/00 , H04N21/44 , G06V20/40
Abstract: The disclosed computer-implemented method may include accessing media segments that correspond to respective media items. At least one of the media segments may be divided into discrete video shots. The method may also include matching the discrete video shots in the media segments to corresponding video shots in the corresponding media items according to various matching factors. The method may further include generating a relative similarity score between the matched video shots in the media segments and the corresponding video shots in the media items, and training a machine learning model to automatically identify video shots in the media items according to the generated relative similarity score between matched video shots. Various other methods, systems, and computer-readable media are also disclosed.
-
公开(公告)号:US20250157187A1
公开(公告)日:2025-05-15
申请号:US18506881
申请日:2023-11-10
Applicant: Netflix, Inc.
Inventor: Aneesh Vartakavi , Dong Liu , Benjamin Eliot Klein
Abstract: A computer-implemented method may include accessing an image associated with a media item and identifying an association between the accessed image and an image take fraction that indicates how well the accessed image correlates to views of the associated media item. Then, based on the identified association between the accessed media item image and the corresponding image take fraction, the method may include training a machine learning (ML) model to predict which images will optimally correlate to views of the associated media item. The method may further include accessing an unprocessed image associated with a new media item that has not been processed by the trained ML model and implementing the trained ML model to predict an image take fraction for the unprocessed image to indicate how well the unprocessed image will correlate to views of the new, unprocessed media item. Various other methods, systems, and computer-readable media are also disclosed.
-
-
-
-
-
-
-