公开(公告)号：US09330171B1

公开(公告)日：2016-05-03

申请号：US14161146

申请日：2014-01-22

Applicant: Google Inc.

Inventor： Sanketh Shetty ,  Andrej Karpathy ,  George Dan Toderici

IPC: G06K9/64 ,  G06F17/30 ,  G06T3/00

CPC classification number: G06F17/30784 ,  G06F17/30247 ,  G06K9/00744 ,  G06K9/64 ,  G06T3/00

Abstract: A method includes receiving, by a processing device of a content sharing platform, a video content, selecting at least one video frame from the video content, subsampling the at least one video frame to generate a first representation of the at least one video frame, selecting a sub-region of the at least one video frame to generate a second representation of the at least one video frame, and applying a convolutional neuron network to the first and second representations of the at least one video frame to generate an annotation for the video content.

Abstract translation: 一种方法包括：通过内容共享平台的处理设备接收视频内容，从视频内容中选择至少一个视频帧，对所述至少一个视频帧进行二次采样以生成所述至少一个视频帧的第一表示， 选择所述至少一个视频帧的子区域以生成所述至少一个视频帧的第二表示，以及将卷积神经网络应用于所述至少一个视频帧的所述第一和第二表示，以生成所述至少一个视频帧的注释 视频内容