IP Library Granted Patent US 12,670,676
Granted Patent B2
US 12,670,676 · App. 18/692,158 · Granted Jun 30, 2026

Method and device for facilitating a privacy-aware representation in a system

Inventors: Konstantinos Vandikas (Solna, SE); Daniel Lindström (Luleå, SE); Lackis Eleftheriadis (Valbo, SE); Athanasios Karapantelakis (Solna, SE)
Assignee: TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
G06T19/006G06F21/6245
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,670,676
App. No.
18/692,158
Granted
Jun 30, 2026
Kind
B2
Abstract

Embodiments herein disclose a method for facilitating a privacy-aware representation in a system. The method comprises determining using a content analyzer a layout of a scene. Thereafter, one or more objects in the scene is identified using the content analyzer to define a relationship between the objects. A privacy status to be tagged to the one or more identified objects is inferred by using a machine learning model. At least one object is processed based on the privacy status inferred. A privacy-aware representation of the scene is rendered, wherein the privacy-aware representation displays the scene with at least one processed object.

Claims (35)

1 . A computer-implemented method for facilitating a privacy-aware representation in a system, the method comprising:

determining, using a content analyzer, a layout of a scene, wherein the scene is received through a user input device and the scene comprises a set of objects with varying privacy requirements;

identifying, using the content analyzer, at least a first object in the scene and a second object in the scene;

determining a relationship between the first object and the second object;

inputting to a machine learning model relationship information indicating the determined relationship between the first object and second object;

after inputting the relationship information to the machine learning model, obtaining from the machine learning model privacy information indicating an inferred privacy status for the first object;

processing the first object based on the inferred privacy status to produce a first processed object; and

rendering a privacy-aware representation of the scene, wherein the privacy-aware representation displays the scene with at least the first processed object.

2 . The method of claim 1 , wherein processing the first object comprises one of: partially concealing the first object, completely concealing the first object, transforming the first object, or transforming an area around the first object.

3 . The method of claim 1 , wherein the method further comprises:

obtaining a receiver's preferences about the first processed object from the privacy-aware representation through a point of interest module.

4 . The method of claim 3 , wherein the point of interest module comprises one or more sensors to capture the receiver's eyes gaze, tactile movements, or touch.

5 . The method of claim 3 , wherein the method further comprises:

providing the receiver's preferences to a sender;

receiving an input from the sender in response to the receiver's preferences; and

defining a user action based on the input from the sender, by the machine learning model to process the first object.

6 . The method of claim 1 , wherein the method further comprises configuring the machine learning model in a training phase with the first object as input to define a state of the first object, wherein the state indicates an existing privacy status associated with the first object.

7 . The method of claim 1 , wherein the method further comprises configuring the machine learning model in a training phase to predict an action to process the first object.

8 . The method of claim 1 , wherein the method further comprises configuring the machine learning model in a training phase to iteratively update the state and action associated with the first object based on the user action.

9 . The method of claim 1 , wherein the method further comprises configuring the machine learning model to calculate a reward, r, which enables an agent to predict an action similar to a user action by minimizing or maximizing a difference between a predicted action and the user action using a weight factor per object.

10 . The method of claim 1 , wherein the machine learning model is trained using one a reinforcement learning algorithm, an unsupervised machine learning, and a Q-learning algorithm.

11 . The method of claim 1 , wherein the scene comprises at least one of an image content, audio-visual content and an audio content.

12 . The method of claim 1 , wherein the user input device is one of a camera, a speaker, a headphone, a network node, a tactile device or any device capable of controlling a sensory actuator.

13 . The method of claim 1 , wherein the system is at least one of a wearable device, a hand-held device, a distributed computing system, or computing device configured to render mixed reality, augmented reality, extended reality, or virtual reality.

14 . A non-transitory computer readable storage medium storing a computer program comprising instructions which when executed by a processing unit of the system causes the system to perform the method of claim 1 .

15 . A system for facilitating a privacy-aware representation, the system comprising:

memory; and

processing circuitry, wherein the system is configured to perform a method comprising:

determining, using a content analyzer, a layout of a scene, wherein the scene is received through a user input device and the scene comprises a set of objects with varying privacy requirements;

identifying, using the content analyzer, at least a first object in the scene and a second object in the scene;

determining a relationship between the first object and the second object;

inputting to a machine learning model relationship information indicating the determined relationship between the first object and second object;

after inputting the relationship information to the machine learning model, obtaining from the machine learning model privacy information indicating an inferred privacy status for the first object;

processing the first object based on the inferred privacy status to produce a first processed object; and

rendering a privacy-aware representation of the scene, wherein the privacy-aware representation displays the scene with at least the first processed object.