In-call experience enhancement for assistant systems

Patent metadata

Field	Value
Publication number	US-12182883-B2
Application number	US-202117500679-A
Country	US
Kind code	B2
Filing date	Oct 13, 2021
Priority date	Oct 18, 2019
Publication date	Dec 31, 2024
Grant date	Dec 31, 2024

Abstract

In one embodiment, a method includes establishing a video call between a plurality of client systems, wherein access to an assistant system is persistently maintained during the video call, receiving, from a first client system of the plurality of client systems, a request by a first user to be performed by the assistant system during the video call, wherein the request references one or more activities associated with one or more users associated with the plurality of client systems, analyzing, by a context engine of the assistant system, images of a scene of the video call to identify the one or more activities within the scene, instructing the assistant system to execute the request based on the identified one or more activities, and sending, to one or more of the plurality of client systems, a response to the request while maintaining the video call between the plurality of client systems.

First claim

What is claimed is:

1. A method comprising, by one or more computing systems:
establishing a video call between a plurality of client systems, wherein access to an assistant system is persistently maintained during the video call;
receiving, from a first client system of the plurality of client systems, a request by a first user to be performed by the assistant system during the video call, wherein the request references one or more activities to be performed after the request and is associated with one or more users associated with the plurality of client systems, wherein the one or more activities referenced in the request are physical actions performed by the one or more users after the request is made;
analyzing, by a context engine of the assistant system after receiving the request from the first user, images of a scene of the video call to identify within the scene the one or more activities referenced in the request;
instructing the assistant system to execute the request responsive to the identification of one or more of the activities referenced in the request; and
sending, to one or more of the plurality of client systems, a response to the request while maintaining the video call between the plurality of client systems.

2. The method of claim 1 , wherein the request by the first user further references an instruction to perform a virtual activity with respect to one or more of the activities.

3. The method of claim 1 , wherein the request by the first user further references an instruction to identify within the scene one or more objects or one or more users with respect to one or more of the activities referenced in the request.

4. The method of claim 1 , wherein analyzing images of the scene of the video call to identify within the scene the one or more activities referenced in the request comprises:
identifying user activity of one or more users of the plurality of client systems, wherein the identified user activity is identified as being equivalent to one of the one or more activities referenced in the request based on a taxonomy of activity classes.

5. The method of claim 1 , further comprising:
accessing, from the context engine of the assistant system, context data associated with the video call, wherein the context data indicates properties of a scene of the video call to identify one or more activities or objects within the scene.

6. The method of claim 5 , wherein the context data comprises identifications of one or more objects within the scene of the video call, and wherein the request by the first user further references one or more of the identified objects.

7. The method of claim 5 , wherein the context data comprises identifications of one or more users within the scene of the video call, and wherein the request by the first user further references one or more of the identified users.

8. The method of claim 5 , wherein the context data comprises location information of one or more objects or one or more users within the scene of the video call, and wherein the request by the first user further references a location of one or more of the objects or users.

9. The method of claim 5 , wherein the context data comprises information indicating one or more context changes with respect to one or more objects within the scene of the video call, and wherein the request by the first user further references context of one or more of the objects.

10. The method of claim 5 , wherein the context data comprises information indicating context changes with respect to one or more users within the scene of the video call, and wherein the request by the first user further references context of one or more users.

11. The method of claim 5 , wherein the context data comprises information indicating context changes with respect to the location of one or more objects or one or more users within the scene of the video call, and wherein the request by the first user further references a location of one or more of the objects or users.

12. The method of claim 1 , wherein the assistant system comprises a scene understanding engine.

13. The method of claim 12 , further comprising:
accessing, from the scene understanding engine of the assistant system, relationship data associated with the scene of the video call, wherein the relationship data indicates relationships between one or more users or one or more objects within the scene of the video call,
wherein the request by the first user further references a relationship between one or more of the objects or users.

14. The method of claim 12 , further comprising:
determining the request by the first user further references a particular type of relationship data; and
activating the scene understanding engine in response to determining the request references the particular type of relationship data, wherein the scene understanding engine analyzes the scene of the video call to generate relationship data of the particular type of relationship data referenced in the request.

15. The method of claim 14 , wherein the scene understanding engine generates the relationship data in real time in response to being activated.

16. The method of claim 14 , further comprising:
deactivating the scene understanding engine after the relationship data has been generated.

17. The method of claim 12 , wherein the scene understanding engine accesses the relationship data in response to a subsequent request by the first user to be performed by the assistant system during the video call.

18. The method of claim 12 , further comprising:
activating the scene understanding engine in response to determining the request references a particular type of relationship, wherein the scene understanding engine analyzes the scene of the video call to generate relationship data of the particular type of relationship referenced in the request;
identifying one or more changes between the particular type of relationship in the scene of the video call; and
in response to determining the request references changes between the particular type of relationship in the scene of the video call, instructing the assistant system to shift the scene of the video call.

19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
establish a video call between a plurality of client systems, wherein access to an assistant system is persistently maintained during the video call;
receive, from a first client system of the plurality of client systems, a request by a first user to be performed by the assistant system during the video call, wherein the request references one or more activities to be performed after the request and is associated with one or more users associated with the plurality of client systems, wherein the one or more activities are physical actions performed by the one or more users after the request is made;
analyze, by a context engine of the assistant system after receiving the request from the first user, images of a scene of the video call to identify within the scene the one or more activities referenced in the request;
instruct the assistant system to execute the request responsive to the identification of one or more of the activities referenced in the request; and
send, to one or more of the plurality of client systems, a response to the request while maintaining the video call between the plurality of client systems.

20. A system comprising one or more processors and one or more computer-readable non-transitory storage medi

Assignees

• Meta Platforms Tech Llc

Inventors

• Liu Xiaohu
• Liu Baiyang
• Subba Rajen

Classifications

• G06Q10/40 Primary

  Business processes related to social networking or social networking services · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/098

  Distributed learning, e.g. federated learning · CPC title

• G06Q10/1093

  Calendar-based scheduling for persons or groups · CPC title

• G10L2015/0631

  Creating reference templates; Clustering · CPC title