公开(公告)号：US10319410B1

公开(公告)日：2019-06-11

申请号：US14977134

申请日：2015-12-21

Applicant: Amazon Technologies, Inc.

Inventor： Matthew Alan Townsend ,  Peter Van Tuyl Bentley ,  Gabrielle Halberg ,  Bryan Harris ,  Geoff Hunter Donaldson

IPC: H04N5/93 ,  H04N9/80 ,  G11B27/034 ,  G11B27/30 ,  G11B27/10

Abstract: A system and method is provided for generating an edit decision list used to render a video summarization. The system may determine a structure indicating a pacing and a sequence of video clips included in the video summarization. The system may determine weighted priority metrics for a plurality of video clips, may identify first video clips having a weighted priority metric above a first threshold and may include the first video clips in the video summarization. The system may determine similarity scores for the plurality of video clips, identify second video clips having a weighted priority metric below the first threshold but above a second threshold and a similarity score below a third threshold and may include a portion of the second video clips in the video summarization. The system may determine layouts, apply special effects and transitions and synchronize music with the video summarization based on a theme and annotation data.

公开(公告)号：US10582125B1

公开(公告)日：2020-03-03

申请号：US14727782

申请日：2015-06-01

Applicant: Amazon Technologies, Inc.

Inventor： Ross David Roessler ,  Matthew Alan Townsend ,  Yinfei Yang ,  Jim Oommen Thomas ,  Deon Poncini ,  William Evan Welbourne ,  Geoff Hunter Donaldson ,  Paul Aksenti Savastinuk ,  Cheng-Hao Kuo

IPC: G03B37/04 ,  H04N5/232 ,  H04N5/225 ,  G06T7/00

Abstract: A video capture device may include multiple cameras that simultaneously capture video data. The video capture device and/or one or more remote computing resources may stitch the video data captured by the multiple cameras to generate stitched video data that corresponds to 360° video. The remote computing resources may apply one or more algorithms to the stitched video data to identify one or more frames that depict content that is likely to be of interest to a user. The video capture device and/or the remote computing resources may generate one or more images from the one or more frames, and may send the one or more images to the user.