Selecting spectator viewpoints in volumetric video presentations of live events

A non-transitory computer-readable storage medium storing instructions which, when executed by a processor, cause the processor to perform operations, the operations comprising:

• receiving a plurality of video streams depicting a live event occurring in a venue, wherein the plurality of video streams are provided by a plurality of cameras which are geographically distributed within the venue;
• identifying an initial position of a target that is present in the venue, based on an analysis of the plurality of video streams;
• compositing the plurality of video streams to produce a volumetric video traversal of the live event, wherein the first volumetric video traversal comprises a first continuous sequence of viewpoints of the live event that follows the target through the venue;
• predicting a future position of the target in the venue at a future point in time, based in part on a current position of the target;
• predicting an occurrence of a future event based on the future position of the target;
• and
• sending, in response to the predicting the occurrence of the future event, an alert to a display device that is streaming a second volumetric video traversal of the live event comprising a second continuous sequence of viewpoints of the live event that is different from the first continuous sequence of viewpoints of the live event, wherein the alert changes an appearance of the target in the second volumetric video traversal of the live event.

A system comprising:

• a processor deployed in a telecommunication service provider network;
• and
• a non-transitory computer-readable medium storing instructions which, when executed by the processor, cause the processor to perform operations, the operations comprising:
• receiving a plurality of video streams depicting a live event occurring in a venue, wherein the plurality of video streams are provided to the processor by a plurality of cameras which are geographically distributed within the venue;
• identifying an initial position of a target that is present in the venue, based on an analysis of the plurality of video streams;
• compositing the plurality of video streams to produce a volumetric video traversal of the live event, wherein the first volumetric video traversal comprises a first continuous sequence of viewpoints of the live event that follows the target through the venue;
• predicting a future position of the target in the venue at a future point in time, based in part on a current position of the target;
• predicting an occurrence of a future event based on the future position of the target;
• and
• sending, in response to the predicting the occurrence of the future event, an alert to a display device that is streaming a second volumetric video traversal of the live event comprising a second continuous sequence of viewpoints of the live event that is different from the first continuous sequence of viewpoints of the live event, wherein the alert changes an appearance of the target in the second volumetric video traversal of the live event.