Machine learning and distributed processing for creating avatars while watching video content

1 . A method for generating an avatar while watching video content comprising:

receiving a request to create the avatar based on the video content;

in response to receiving the request to create the avatar based on the video content, determining one or more processing resources to use to create the avatar based on an output of a machine learning engine, wherein the one or more processing resources include a cloud-based processing resource, an edge-based processing resource, and a local processing resource;

allocating avatar processing operations to the one or more processing resources to create the avatar based on the output from the machine learning engine;

creating the avatar using the one or more processing resources, wherein creating the avatar further comprises:

requesting a character feature not displayed in the video content from different video content or images available at the cloud-based processing resource,

providing the character feature from the different video content or images to the edge-based processing resource, and

using the edge-based processing resource to create the avatar using the character feature from the different video content or images; and

storing the avatar in an avatar database.

2 . The method of claim 1 , wherein the output of the machine learning engine is based on a complexity of the avatar and whether existing different viewing angle images are available to complete a full model of the avatar.

3 . The method of claim 1 , wherein the output of the machine learning engine is based on a plurality of inputs including one or more of a subscription type, prior avatars created for a subscriber, an amount of available processing bandwidth, processing bandwidth required for creating the avatar, type of video content being watched, time of day, an amount of latency, and an estimated time to process the avatar.

4 . The method of claim 1 , wherein the output of the machine learning engine corresponds with at least one of:

creating the avatar using the cloud-based processing resource;

creating the avatar using the edge-based processing resource;

creating the avatar using a combination of the cloud-based processing resource and the edge-based processing resource;

creating the avatar using a combination of the edge-based processing resource and the local processing resource; and

creating the avatar using a combination of the cloud-based processing resource and the local processing resource.

5 . The method of claim 1 , wherein the output of the machine learning engine corresponds with creating the avatar using the edge-based processing resource to reduce an amount of latency while creating or modifying the avatar.

6 . The method of claim 1 , wherein creating the avatar using the one or more processing resources includes generating a three-dimensional (3D) model of the one or more characters or objects.

7 . The method of claim 1 , further comprising receiving a request to immerse the avatar into a virtual environment.

8 . The method of claim 7 , further comprising sending a request to the machine learning engine to identify one or more of the processing resources to process immersing the avatar in the virtual environment.

9 . The method of claim 1 , further comprising using the machine learning engine to determine character features of a character selected for the avatar that were not displayed in the video content.

10 . The method of claim 1 , further comprising using the machine learning engine to determine which of the one or more processing resources to create the avatar based on a type of the video content being watched, wherein the type of video content includes linear video content, video-on-demand (VOD) content, broadcast video content, subscribed to video content, and unsubscribed to video content.

11 . A system for generating an avatar while watching video content comprising:

a cloud-based processing resource;

an edge-based processing resource;

a local processing resource;

a machine learning engine; and

an avatar generation platform configured to:

receive a request to create the avatar based on the video content;

in response to receipt of the request to create the avatar from the video content, send a request to the machine learning engine to identify one or more processing resources to create the avatar, wherein the one or more processing resources include the cloud-based processing resource, the edge-based processing resource, and the local processing resource;

receive an output from the machine learning engine that allocates avatar processing operations to the one or more processing resources to create the avatar;

create the avatar using the one or more processing resources, wherein the avatar generation platform being configured to create the avatar comprises the avatar generation platform being configured to:

request character features not displayed in the video content from different video content or images available at the cloud-based processing resource,

provide the character features from the different video content or images to the edge-based processing resource, and

use the edge-based processing resource to create the avatar using the character features from the different video content or images; and

store the avatar.

12 . The system of claim 11 , wherein the output of the machine learning engine is based on a plurality of inputs including one or more of a subscription type, prior avatars created for a subscriber, an amount of available processing bandwidth, processing bandwidth required for creating the avatar, type of video content being watched, time of day, an amount of latency, and an estimated time to process the avatar.

13 . The system of claim 11 , wherein the output of the machine learning engine corresponds with at least one of:

creation of the avatar using the cloud-based processing resource;

creation of the avatar using the edge-based processing resource;

creation of the avatar using a combination of the cloud-based processing resource and the edge-based processing resource;

creation of the avatar using a combination of the edge-based processing resource and the local processing resource; and

creation of the avatar using a combination of the cloud-based processing resource and the local processing resource.

14 . The system of claim 11 , further configured to, in response to a request to immerse the avatar, send a request to the machine learning engine to identify one or more of the processing resources to process immersion of the avatar in the virtual environment.

15 . A non-transitory computer readable storage medium that includes instructions which, when executed, cause a processor to:

receive a request to create the avatar based on the video content;

in response to receipt of the request to create the avatar from the video content, send a request to a machine learning engine to identify one or more processing resources to create the avatar, wherein the one or more processing resources include a cloud-based processing resource, an edge-based processing resource, and a local processing resource;

receive an output from the machine learning engine that allocates avatar processing operations to the one or more processing resources to create the avatar;

create the avatar using the one or more processing resources, wherein the instructions causing the processor to create the avatar further comprises the instructions causing the processor to:

request character features not displayed in the video content from different video content or images available at the cloud-based processing resource,

provide the character features from the different video content or images to the edge-based processing resource, and

use the edge-based processing resource to create the avatar using the character features from the different video content or images; and

store the avatar in a computer storage device.

16 . The non-transitory computer readable storage medium of claim 15 , wherein the output of the machine learning engine is based on a plurality of inputs including one or more of a subscription type, prior avatars created for a subscriber, an amount of available processing bandwidth, processing bandwidth required for creating the avatar, type of video content being watched, time of day, an amount of latency, and an estimated time to process the avatar.

17 . The non-transitory computer readable storage medium of claim 15 , wherein the output of the machine learning engine corresponds with at least one of:

creation of the avatar using the cloud-based processing resource;

creation of the avatar using the edge-based processing resource;

creation of the avatar using a combination of the cloud-based processing resource and the edge-based processing resource;

creation of the avatar using a combination of the edge-based processing resource and the local processing resource; and

creation of the avatar using a combination of the cloud-based processing resource and the local processing resource.

18 . The non-transitory computer readable storage medium of claim 15 , wherein creating the avatar using the one or more processing resources includes generating a three-dimensional (3D) model of the one or more characters or objects.

19 . The non-transitory computer readable storage medium of claim 15 , further comprising receiving a request to immerse the avatar into a virtual environment.

20 . The non-transitory computer readable storage medium of claim 15 , further comprising sending a request to the machine learning engine to identify one or more of the processing resources to process immersing the avatar in the virtual environment.