-
公开(公告)号：US10826853B1
公开(公告)日：2020-11-03
申请号：US15916741
申请日：2018-03-09
Applicant: Facebook, Inc.
Inventor： Ryan Michael O'Rourke , Yichen Wang , Nathan Andrew Sharp
IPC: G06F15/16 , H04L12/58 , G06Q50/00 , G06F16/9535
Abstract: Systems, methods, and non-transitory computer-readable media can determine at least one media content item being posted through a social networking system, wherein the media content item is able to be shared through one or more non-ephemeral content feeds, an ephemeral story feed, or with a favorites list of users; determine that the media content item is being shared with the favorites list of users; determine users included in the favorites list; and distribute the media content item to users included in the favorites list.
-
公开(公告)号：US10824903B2
公开(公告)日：2020-11-03
申请号：US16550774
申请日：2019-08-26
Applicant: Facebook, Inc.
Inventor： Michael Fabien Mathieu , Camille Couprie , Yann Andre Le Cun
Abstract: In one embodiment, a method includes receiving a plurality of input frames of a video sequence associated with a time t, predicting, using a convolutional network, one or more future frames of the video sequence from the plurality of input frames, wherein the convolutional network is trained with randomly selected temporal sequences of a n×m grid of pixels from the plurality of input frames exhibiting a threshold of optical flow. In addition, the training may comprise randomly selecting temporal sequences of a n×m grid of pixels from the plurality of input frames exhibiting a threshold of optical flow.
-
公开(公告)号：US20200341716A1
公开(公告)日：2020-10-29
申请号：US16883819
申请日：2020-05-26
Applicant: Facebook, Inc.
Inventor： Joshua Alexander Miller , Leo Litterello Mancini , Michael Slater
IPC: G06F3/14 , H04L29/06 , H04N21/43 , H04N21/4788 , H04N21/482 , H04L29/08 , H04L12/58
Abstract: One or more embodiments of the disclosure provide systems and methods for providing media presentations to users of a media presentation system. A media presentation generally includes a plurality of media segments provided by multiple users of the media presentation system. In one or more embodiments, a user of the media presentation system may share a media presentation with a co-user. The media presentation system can enable the co-user, if authorized by the user, to contribute (e.g., add a media segment) to a media presentation shared with the co-user.
-
公开(公告)号：US10819949B1
公开(公告)日：2020-10-27
申请号：US16442928
申请日：2019-06-17
Applicant: Facebook, Inc.
Inventor： Zhengping Zuo , Stephane Taine , Brendan Benjamin Aronoff
Abstract: Exemplary embodiments relate to techniques for facilitating a connection between two users in a call by providing an “answering window” when a user calls. When a caller places a call to a callee, the call invitation may remain open for a window of time during which the caller remains available. The callee may opt to answer the call at any time during the window. During the window, the caller may be placed in a lobby and may share audio and/or video. Accepting the call during the answering window may place the callee in the lobby or may directly convert the interaction into an active call. The callee may request that the window can be altered, or may signal their intention to answer the call at a particular time.
-
公开(公告)号：US10817783B1
公开(公告)日：2020-10-27
申请号：US16868936
申请日：2020-05-07
Applicant: Facebook, Inc.
Inventor： Nadav Rotem , Abdulkadir Utku Diril , Mikhail Smelyanskiy , Jong Soo Park , Christopher Dewan
Abstract: The disclosed computer-implemented method for efficiently updating neural networks may include (i) identifying a neural network that comprises sets of interconnected nodes represented at least in part by a plurality of matrices and that is trained on a training computing device and executes on at least one endpoint device, (ii) constraining a training session for the neural network to reduce the size in memory of the difference between the previous values of the matrices prior to the training session and the new values of the matrices after the training session, (iii) creating a delta update for the neural network that describes the difference between the previous values and the new values, and (iv) updating the neural network on the endpoint device to the new state by sending the delta update from the training computing device to the endpoint computing device. Various other methods, systems, and computer-readable media are also disclosed.
-
Patent Agency Ranking
公开(公告)号：US20200336509A1
公开(公告)日：2020-10-22
申请号：US16388058
申请日：2019-04-18
Applicant: Facebook, Inc.
Inventor： Haotian Wang , Komal Kapoor , Gaurav Singh Thakur
IPC: H04L29/06 , G06N20/20 , G06F16/951 , H04L29/08
Abstract: An online system maintains pages and accesses a graph of nodes representing the pages. Each node is labeled to indicate that a corresponding page is for a real-world entity, an imposter of the real-world entity, or a derived entity complying with or violating a policy. The online system retrieves machine-learning models, each of which is trained based on labels for a set of the nodes and features of corresponding pages. A first model predicts whether a page is for a derived entity based on features of the page. Responsive to predicting the page is not for a derived entity, a second model predicts whether the page is for a real-world entity or an imposter based on features of the page. Responsive to predicting the page is for a derived entity, a third model predicts whether the derived entity complies with or violates the policy based on features of the page.
-
公开(公告)号：US20200334318A1
公开(公告)日：2020-10-22
申请号：US16921424
申请日：2020-07-06
Applicant: Facebook, Inc.
Inventor： Xiaojun Liang , Varun Kacholia , Maher Afif Saba , Mahmud Sami Tas , Gang Liu
IPC: G06F16/957 , G06F3/0482
Abstract: Systems, methods, and non-transitory computer-readable media can provide a content feed through a display interface. The content feed can include at least a snippet of a first content item that is associated with a preview time threshold. A determination can be made that the snippet of the first content item has been positioned within a viewport region of the display interface for a duration of time that satisfies the preview time threshold. Data associated with the first content item can be obtained from a content distribution system. The data associated with the first content item can be stored.
-
公开(公告)号：US10812437B1
公开(公告)日：2020-10-20
申请号：US15943639
申请日：2018-04-02
Applicant: Facebook, Inc.
Inventor： Xiangyang Liu , Lu Zhang
Abstract: A messaging system receives a contact list from a client device. The messaging system receives information about ephemeral content associated with contacts in the contact list and information about the active now status of contacts in the contact list. Based at least in part on the ephemeral content information and the active now status of the contacts in the contact list, the messaging system generates scores for contacts in the contact list using a unified scoring model. The unified scoring model receives as input information about multiple features associated with the user of the messaging system and the contact in the contact list and generates a score representing a probability that the user will interact with the contact upon being presented with the contact. The messaging system ranks the contacts in the list based on the generated scores and transmits the contact list to the client device for display in an activity unit.
-
公开(公告)号：US10809974B2
公开(公告)日：2020-10-20
申请号：US15727132
申请日：2017-10-06
Applicant: Facebook, Inc.
Inventor： Jenny Yuen , David Harry Garcia
Abstract: In one embodiment, a method includes receiving, from a client system of a first user of an online social network, voice input of the first user; receiving, from the client system, information indicating a current geo-location of the first user; parsing the voice input to identify one or more commands in the voice input; determining a data object of the online social network corresponding to the current geo-location of the first user and a command of the one or more commands associated with the current geo-location of the first user; and executing the command associated with the current geo-location of the first user on the online social network, wherein the executed command operates on the data object corresponding to the current geo-location of the first user.
-
公开(公告)号：US10809069B2
公开(公告)日：2020-10-20
申请号：US16679091
申请日：2019-11-08
Applicant: Facebook, Inc.
Inventor： Kristopher T. Frazier , Brian F. Roberts , Donald H. Relyea
IPC: G01C21/26 , H04W4/02 , H04L29/08 , G06F16/29 , G06F16/487 , H04L12/58 , G06F9/445 , G06F9/54 , G06Q30/02
Abstract: In one embodiment, a method includes receiving a content instance from a first device, wherein the content instance is associated with a geographic location and a period of time; determining that a second device is associated with the geographic location and the period of time associated with the content instance, wherein the determination is based on the second device being within a predefined proximity of the geographic location associated with the content instance during the period of time associated with the content instance; and enabling, in response to determining that the second device is associated with the geographic location and the period of time associated with the content instance, the content instance to be accessible by the second device.
-
-
-
-
-
-
-
-
-
Search Results
8633
records
(took 0.033 seconds)
