公开(公告)号：US20180227632A1

公开(公告)日：2018-08-09

申请号：US15865590

申请日：2018-01-09

Applicant: Facebook, Inc.

Inventor： Clare Rubin ,  Chen Sun ,  Maria Angelidou

IPC: H04N21/458 ,  H04N21/2187 ,  H04N21/81 ,  G06Q30/02 ,  G06Q50/00

Abstract: In one embodiment, a method includes providing an interface that permits a first user to broadcast a live video to one or more second users. The method also includes determining that at least one predetermined condition for taking a commercial break during the live video has been met. The method also includes, after determining that at least one of the predetermined conditions has been met, displaying a selectable option within the interface that permits the first user to begin the commercial break. The method also includes, after receiving an indication that the first user has selected the selectable option to begin the commercial break, beginning the commercial break by ceasing to display the live video to the one or more second users after a predetermined amount of time and displaying one or more commercials to the one or more second users during the commercial break.

公开(公告)号：US20170046613A1

公开(公告)日：2017-02-16

申请号：US15091490

申请日：2016-04-05

Applicant: Facebook, Inc.

Inventor： Balamanohar Paluri ,  Lubomir Bourdev ,  Ronan Stéfan Collobert ,  Chen Sun

IPC: G06N3/04 ,  G06N3/08

CPC classification number: G06N3/0454

Abstract: Systems, methods, and non-transitory computer-readable media can obtain a content item to be evaluated by a set of cascaded convolutional neural networks, the set of cascaded convolutional neural networks including at least a first convolutional neural network (CNN) and a second CNN. The content item can be provided to the first CNN as input, wherein an output of the first CNN includes data describing at least one region of interest in the content item and at least one first concept corresponding to the region of interest. The output of the first CNN can be provided to the second CNN as input, wherein an output of the second CNN includes data describing at least one second concept corresponding to the region of interest, the second concept being more accurate than the first concept.

Abstract translation: 系统，方法和非暂时计算机可读介质可以获得要由一组级联卷积神经网络评估的内容项，所述级联卷积神经网络集合包括至少第一卷积神经网络（CNN）和第二卷积神经网络 CNN。 内容项可以作为输入提供给第一CNN，其中第一CNN的输出包括描述内容项中的至少一个感兴趣区域的数据和对应于感兴趣区域的至少一个第一概念。 第一CNN的输出可以作为输入提供给第二CNN，其中第二CNN的输出包括描述与感兴趣区域对应的至少一个第二概念的数据，第二概念比第一概念更准确。