公开(公告)号：US10061985B2

公开(公告)日：2018-08-28

申请号：US15395511

申请日：2016-12-30

Applicant: Facebook, Inc.

Inventor： Balmanohar Paluri ,  Benoit F. Dumoulin ,  Merlyn Deng ,  Reena Philip ,  Dario Garcia Garcia

IPC: G06K9/46 ,  G06K9/00 ,  H04L12/26 ,  G06K9/62 ,  H04N21/44 ,  H04N21/439 ,  H04N21/4788 ,  H04N21/258 ,  H04N21/45 ,  H04N21/482 ,  H04N21/84 ,  H04L29/08

CPC classification number: G06K9/00744 ,  G06K9/00677 ,  G06K9/00718 ,  G06K9/4604 ,  G06K9/6215 ,  G06K9/623 ,  G06K9/6267 ,  G10L15/26 ,  H04L43/045 ,  H04L67/02 ,  H04L67/306 ,  H04N21/25891 ,  H04N21/4394 ,  H04N21/44008 ,  H04N21/4532 ,  H04N21/4788 ,  H04N21/4826 ,  H04N21/84

Abstract: In one embodiment, a method includes accessing a first feature vector representing a video-content object corresponding to a node in a social graph, wherein the video-content object comprises frames and audio and is associated with text, the first feature vector is based on one or more of the frames; accessing a second feature vector representing the video-content object, wherein the second feature vector is based on at least some of the text; accessing a third feature vector representing the video-content object, wherein the third feature vector is based on one or more portions of the audio; determining a fourth feature vector representing the video-content object, wherein the fourth feature vector is based on a combination of the first, second, and third feature vectors; and determining a context of the video-content object based on the fourth feature vector and social-graph information.

公开(公告)号：US20190005332A1

公开(公告)日：2019-01-03

申请号：US16114059

申请日：2018-08-27

Applicant: Facebook, Inc.

Inventor： Balmanohar Paluri ,  Benoit F. Dumoulin ,  Merlyn Deng ,  Reena Philip ,  Dario Garcia Garcia

IPC: G06K9/00 ,  H04N21/84 ,  H04N21/4788 ,  H04N21/45 ,  H04N21/44 ,  H04N21/439 ,  H04N21/258 ,  H04N21/482 ,  G06K9/62 ,  H04L12/26 ,  G10L15/26 ,  G06K9/46 ,  H04L29/08

Abstract: In one embodiment, a method includes accessing a video-content object, determining a first feature vector representing the video-content object using a first recognition module of a first type based on an object in the video-content object, and determining a second feature vector representing the video-content object using a second recognition module of a second type based on the first feature vector. The first type is different from the second type. The method also includes determining a context of the video-content object based on the second feature vector.

公开(公告)号：US20180189570A1

公开(公告)日：2018-07-05

申请号：US15395511

申请日：2016-12-30

Applicant: Facebook, Inc.

Inventor： Balmanohar Paluri ,  Benoit F. Dumoulin ,  Merlyn Deng ,  Reena Philip ,  Dario Garcia Garcia

IPC: G06K9/00 ,  H04L12/26 ,  G06K9/46 ,  G06K9/62 ,  H04N21/44 ,  H04N21/439 ,  H04N21/4788 ,  H04N21/258 ,  H04N21/45 ,  H04N21/482 ,  H04N21/84

CPC classification number: G06K9/00744 ,  G06K9/00718 ,  G06K9/4604 ,  G06K9/623 ,  H04L43/045 ,  H04L67/306 ,  H04N21/25891 ,  H04N21/4394 ,  H04N21/44008 ,  H04N21/4532 ,  H04N21/4788 ,  H04N21/4826 ,  H04N21/84

Abstract: In one embodiment, a method includes accessing a first feature vector representing a video-content object corresponding to a node in a social graph, wherein the video-content object comprises frames and audio and is associated with text, the first feature vector is based on one or more of the frames; accessing a second feature vector representing the video-content object, wherein the second feature vector is based on at least some of the text; accessing a third feature vector representing the video-content object, wherein the third feature vector is based on one or more portions of the audio; determining a fourth feature vector representing the video-content object, wherein the fourth feature vector is based on a combination of the first, second, and third feature vectors; and determining a context of the video-content object based on the fourth feature vector and social-graph information.