GENERATING VIDEOS WITH MULTIPLE VIEWPOINTS

摘要

Videos of an event may be captured by camera (e.g., camera devices). The videos may be processed to generated virtual videos that provide different viewpoints of the event. The videos and virtual videos may be analyzed to identify panning points in the videos. A user may pause the video at the identified panning points and may be allowed to pan around the event and/or event location during the identified panning points in the videos and/or virtual videos. The pan around view may be provided by a pan around video generated based on the videos and/or the virtual videos. A user may resume viewing one of the videos and/or virtual videos after pausing playback of the pan around video. A user may also pan around an event and/or event location while the videos and/or virtual videos are playing during the panning points.

主权项

1. A method comprising:
receiving a plurality of videos of an event, wherein the plurality of videos are associated with a plurality of viewpoints of the event; receiving one or more virtual videos of the event, wherein the one or more virtual videos are associated with one or more of the plurality of viewpoints of the event; identifying a plurality of panning points, wherein each panning point in the plurality of panning points is associated with one video from the plurality of videos or one virtual video from the one or more virtual videos; and generating one or more pan around videos of the event based on the plurality of panning points.