US-12627859-B2 - Systems and methods for recommending content items based on an identified posture

Abstract

Systems and methods are provided for generating a content item recommendation based on an identified posture. An input associated with a content item delivery service is received at a computing device. A capture of a user is received, and a digital representation of the user is generated based on the capture of the user. A posture of the user is determined based on the digital representation of the user, and a content item genre is identified based on the determined posture. A content item recommendation that is based on the identified genre is generated and output.

Inventors

• Rupayan Dutta
• Vikram Makam Gupta
• Vishwas Sharadanagar Panchaksharaiah
• Sukanya Agarwal
• Reda Harb

Assignees

• ADEIA GUIDES INC.

Dates

Publication Date

20260512

Application Date

20240910

Claims (20)

1. 1 . A computer-implemented method comprising: receiving, at a computing device, an input requesting to access a content item delivery service, wherein the computing device is in communication with the content item delivery service over a network, wherein the input is received from a user, and wherein each of the computing device and the user is present in an environment; determining, at a first time and based at least in part on data associated with the user, a current posture of the user in the environment; identifying a known posture that most closely corresponds to the current posture of the user in the environment; determining, at a second time later than the first time, that the user has been maintaining the current posture for a period of time spanning from the first time to the second time; determining that the period of time exceeds a threshold period of time; based at least in part on determining that the period of time exceeds the threshold period of time, transmitting, to the content item delivery service over the network, an indication of the known posture; receiving, at the computing device from the content item delivery service, a content item recommendation, wherein the content item recommendation is based at least in part on a content type that is determined based at least in part on the indication of the known posture; and generating for output, based on the identified content type, the content item recommendation.
2. 2 . The method of claim 1 , wherein the data associated with the user is a captured image of the user in the environment, and the current posture of the user is determined based at least in part by processing the captured image.
3. 3 . The method of claim 1 , wherein the data associated with the user is a capture of the user comprising at least one of: an image from a camera, point data from a LiDAR device, output from an ultrasonic sensor, or output from a laser sensor.
4. 4 . The method of claim 1 , further comprising determining an emotion of the user based at least in part on the known posture, wherein the content type is further determined based at least in part on the determined emotion.
5. 5 . The method of claim 1 , further comprising: determining that the current posture has changed and more closely corresponds to a different known posture; identifying, based at least in part on the different known posture, an updated content type; and generating for output, based at least in part on the updated content type, an updated content item recommendation.
6. 6 . The method of claim 1 , wherein: the identifying, based at least in part on the known posture, the content type further comprises identifying, based at least in part on the known posture, weightings for a plurality of content types; and generating the content item recommendation further comprises generating content item recommendations based at least in part on the weightings for the plurality of content types.
7. 7 . The method of claim 1 , wherein identifying the content type further comprises determining the content type via a posture-to-type mapping table, the method further comprising: identifying an interaction level associated with the content item recommendation; determining whether the interaction level is below a threshold value; and, based at least in part on determining that the interaction level is below the threshold value: updating the posture-to-type mapping table to associate a different known posture with the content type on which the content item recommendation has been generated.
8. 8 . The method of claim 1 , wherein: the data associated with the user is further associated with a plurality of users; the method further comprises: generating a probability value associated with each user of the plurality of users, wherein the probability value indicates a probability of each user interacting with the content item delivery service; and wherein: the known posture is identified for the user having a highest probability value associated with them.
9. 9 . The method of claim 1 , wherein the input is spoken input, and the method further comprises: identifying a user profile based at least in part on the spoken input; and associating the known posture with the user profile.
10. 10 . The method of claim 1 , further comprising: receiving, at the computing device, a segment of a content item and an associated manifest file; identifying, via the associated manifest file, an indication to determine the current posture of the user at a playback time of the content item; and wherein: determining the current posture of the user further comprises determining the current posture of the user at the indicated playback time.
11. 11 . A system comprising: a communication port; a memory storing instructions; and control circuitry communicably coupled to the memory and the communication port and configured to execute instructions to: receive, at a computing device, an input requesting to access a content item delivery service, wherein the computing device is in communication with the content item delivery service over a network, wherein the input is received from a user, and wherein each of the computing device and the user is present in an environment; determine, at a first time and based at least in part on data associated with a current posture of the user in the environment; identify a known posture that most closely corresponds to the current posture of the user in the environment; determine, at a second time later than the first time, that the user has been maintaining the current posture for a period of time spanning from the first time to the second time; determine that the period of time exceeds a threshold period of time; based at least in part on determining that the period of time exceeds the threshold period of time, transmit, to the content item delivery service, an indication of the known posture; receive, at the computing device from the content item delivery service, a content item recommendation, wherein the content item recommendation is based at least in part on a content type that is determined based at least in part on the indication of the known posture; and generate for output, based at least in part on the identified content type, the content item recommendation.
12. 12 . The system of claim 11 , wherein the data associated with the user is a captured image of the user in the environment, and the current posture of the user is determined based at least in part by processing the captured image.
13. 13 . The system of claim 11 , wherein the data associated with the user is a capture of the user comprising at least one of: an image from a camera, point data from a LiDAR device, output from an ultrasonic sensor, or output from a laser sensor.
14. 14 . The system of claim 11 , wherein the control circuitry is further configured to: determine an emotion of the user based at least in part on the known posture; and determine the content type based at least in part on the determined emotion.
15. 15 . The system of claim 11 , wherein the control circuitry is further configured to: determine that the current posture has changed and more closely corresponds to a different known posture; identify, based at least in part on the different known posture, an updated content type; and generate for output, based at least in part on the updated content type, an updated content item recommendation.
16. 16 . The system of claim 11 , wherein: the control circuitry configured to identify, based at least in part on the known posture, the content type, is further configured to identify, based at least in part on the known posture, weightings for a plurality of content types; and wherein the control circuitry configured to generate the content item recommendation, is further configured to generate content item recommendations based at least in part on the weightings for the plurality of content types.
17. 17 . The system of claim 11 , wherein the control circuitry is configured to identify the content type by determining the content type via a posture-to-type mapping table, and wherein the control circuitry is further configured to: identify an interaction level associated with the content item recommendation; determine whether the interaction level is below a threshold value; and, based at least in part on determining that the interaction level is below the threshold value: update the posture-to-type mapping table to associate a different known posture with the content type on which the content item recommendation has been generated.
18. 18 . The system of claim 11 , wherein the data associated with the user is further associated with a plurality of users, and wherein the control circuitry is further configured to: generate a probability value associated with each user of the plurality of users, wherein the probability value indicates a probability of each user interacting with the content item delivery service; and wherein: the known posture is identified for the user having a highest probability value associated with them.
19. 19 . The system of claim 11 , wherein the input is spoken input, and wherein the control circuitry is further configured to: identify a user profile based at least in part on the spoken input; and associate the known posture with the user profile.
20. 20 . The system of claim 11 , wherein the control circuitry is further configured to: receive, at the computing device, a segment of a content item and an associated manifest file; identify, via the associated manifest file, an indication to determine the current posture of the user at a playback time of the content item; and wherein: the control circuitry determines the current posture of the user by determining the current posture of the user at the indicated playback time.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a continuation of U.S. patent application Ser. No. 18/383,608, filed Oct. 25, 2023, which is a continuation of U.S. patent application Ser. No. 17/752,311, filed May 24, 2022, now U.S. Pat. No. 11,843,829, the disclosures of which are hereby incorporated by reference herein in their entireties. BACKGROUND The present disclosure is directed towards systems and methods for generating a content item recommendation. In particular, systems and methods are provided herein that generate a content item recommendation based on an identified posture. SUMMARY The proliferation of content item delivery services, including over-the-top (OTT), streaming and video on demand (VOD) services, such as Netflix, Amazon Prime Video, Disney+ and Hulu, affords users of content items more choice than ever before. However, when a user first accesses a content item delivery service, they may not have a clear idea of what they want to view. This issue may be particularly apparent when a user accesses a content item delivery service for the first time, as the user may not be aware of all the content items that are available. In addition, the content item delivery service may have a wide breadth of content items available, which may make it more difficult for the user to find something suitable to consume. This may lead a user to access one content item after another, looking for something to consume. For example, a user may access the first episode of many different series. As the user of the content item delivery service may ultimately discard and/or skip content that they are not interested in consuming, network bandwidth, storage resources and/or processing resources will be wasted during the delivery of additional content that is not relevant. In order to reduce the consumption of these resources, many content item delivery services may recommend content items to a user. In order to make recommendations more pertinent to the user, typically these recommendations may be based on, or generated from, previously consumed content items. However, when a user is accessing a content item delivery service for the first time, there is no consumption history on which to base recommendations. As such, resources may be wasted when a new user is looking for content to consume on a content item delivery service. In addition, at times, a user may prefer, for example, romance content items, and at other times, the user may prefer action content items. As such, simply basing recommendations on previously consumed content items may not give rise to optimal recommendations, and resources may be wasted while a user searches for a particular genre of content item to watch. To overcome these problems, systems and methods are provided herein for generating a content item recommendation. In particular, systems and methods are provided herein that generate a content item recommendation based on an identified posture. Systems and methods are described herein for generating a content item recommendation. In accordance with some aspects of the disclosure, a method is provided for generating a content item recommendation based on an identified posture. An input associated with a content item delivery service is received at a computing device, and a capture of a user is received. A digital representation of the user is generated based on the capture of the user, and a posture of the user is determined based on the digital representation of the user. A content item genre is identified based on the determined posture of the user, and a content item recommendation is generated based on the identified genre. The content item recommendation is output. In an example system, a user opens an OTT application, such as a Netflix application, on a smart television, via a command issued from a remote control. On receiving the command, software running on the smart television opens the OTT application. On opening the OTT application, a camera integrated into the smart television takes a photo of the user. A posture of the user is determined, based on the photo of the user. This may comprise, for example, locating skeletal joints of the user in order to determine a posture of the user. In some examples, the posture of the user is determined locally, at the smart television. In some examples, the user posture may be determined via a trained algorithm. The smart television may comprise, for example, an artificial intelligence accelerator chip, such as a Google Tensor chip, or a Samsung Exynos chip, which may be used in determining the posture of the user. In other examples, the capture of the user is transmitted via a network, such as the internet, to a server. Software running at the server may determine a posture of the user, and an indication of the posture may be transmitted back to the smart television and/or to a program running on another server (or the same server). Based on the identified posture, a genre is id