Method and electronic device for managing private content

Patent metadata

Field	Value
Publication number	US-12482290-B2
Application number	US-202217740531-A
Country	US
Kind code	B2
Filing date	May 10, 2022
Priority date	Jul 31, 2020
Publication date	Nov 25, 2025
Grant date	Nov 25, 2025

Abstract

Embodiments herein provide a method for managing private content by an electronic device ( 100 ). The method includes: displaying content on a screen of the electronic device ( 100 ), wherein the content is stored in a non-secure folder of the electronic device ( 100 ) and extracting a feature(s) of the displayed content to categorize the displayed content using a machine learning (ML) model. Further, the method includes categorizing the displayed content into the private content or public content based on globally objectionable private content and user-specific objectionable private content. Further, the method includes displaying a recommendation on the screen of the electronic device ( 100 ) for transferring the displayed content into a secure folder of the electronic device ( 100 ), detecting an event to transfer the private content into the secure folder of the electronic device ( 100 ), and automatically transferring the private content from the non-secure folder to the secure folder of the electronic device ( 100 ).

First claim

What is claimed is:

1 . A method for managing private content by an electronic device, the method comprising:
displaying, by the electronic device, content on a screen of the electronic device, wherein the content is stored in a non-secure folder of the electronic device; extracting, by the electronic device, at least one feature of the displayed content to categorize the displayed content using at least one Machine Learning (ML) model; detecting, by the electronic device, at least one of a globally objectionable private content and a user-specific objectionable private content; categorizing, by the electronic device, the displayed content into one of the private content and public content based on the at least one of the globally objectionable private content and the user-specific objectionable private content; displaying, by the electronic device, a recommendation on the screen of the electronic device for transferring the displayed content into a secure folder of the electronic device in response to categorizing the displayed content as the private content; detecting, by the electronic device, at least one event to transfer the private content into the secure folder of the electronic device; and automatically transferring, by the electronic device, the private content from the non-secure folder to the secure folder of the electronic device in response to detecting the at least one event, wherein categorizing, by the electronic device, the displayed content into the one of the private content and the public content based on the globally objectionable private content comprises: determining, by the electronic device, whether at least one of a private body part of a human is visible in the displayed content and an objectionable element is present in the displayed content; performing, by the electronic device, at least one of: categorizing the displayed content as the private content in response to determining that the at least one of the private body part of the human is visible in the displayed content and the objectionable element is present in the displayed content; and determining a cumulative confidence score to categorize the displayed content as the at least one of the private content and the public content in response to determining that the at least one of the private body part of the human is not visible in the displayed content and the objectionable element is not present in the displayed content, wherein determining the cumulative confidence score to categorize an image frame of the displayed content as the at least one of the private content and the public content comprises: detecting, by the electronic device, that at least one private body part of the human is not visible in the image frame; determining, by the electronic device, whether multiple human blocks are identified in the image frame; and performing, by the electronic device, at least one of: determining a first threshold value in response to determining that the multiple human blocks are identified in the image frame; and determining a second threshold value using a sensitive image classifier in response to determining that the multiple human blocks are not identified in the image frame; determining, by the electronic device, the cumulative confidence score by adding the first threshold value and the second threshold value, wherein determining the first threshold value comprises: detecting, by the electronic device, a block proximity threshold; determining, by the electronic device, whether the multiple human blocks meet the block proximity threshold; performing, by the electronic device, at least one of: grouping the multiple human blocks into a single block and passing the multiple human blocks to the sensitive image classifier to calculate the first threshold value in response to determining that the multiple human blocks meet the block proximity threshold; and passing individual blocks to the sensitive image classifier to calculate the first threshold value in response to determining that the multiple human blocks do not meet the block proximity threshold.

2 . The method as claimed in claim 1 , wherein the displayed content is categorized as the private content based on the cumulative confidence score being greater than or equal to a specified threshold value and the displayed content is categorized as the public content based on the cumulative confidence score being less than the specified threshold value.

3 . The method as claimed in claim 1 , wherein the private body part of the human comprises at least one of a belly of the human, buttocks of the human, breasts of the human, and genitalia of the human.

4 . The method as claimed in claim 1 , wherein the at least one objectionable element comprises at least one of a cigarette, a beer bottle, a wine bottle, a champagne bottle, a beer glass, a wine glass, a whiskey glass, a cocktail glass, biometric data of a user, medical information of the user, Personally Identifiable Financial Information (PIFI) of the user and unique identifiers of the user.

5 . The method as claimed in claim 1 , wherein categorizing, by the electronic device, the displayed content into the one of the private content and the public content based on the user-specific objectionable private content comprises:
determining, by the electronic device, a user preference for the private content and the public content from a plurality of domains of the electronic device using a multi-modal joint embedding space; storing, by the electronic device, the user preference as at least one template in the electronic device; and correlating, by the electronic device, the at least one extracted feature with the at least one stored template to categorize the displayed content into the one of the private content and the public content.

6 . The method as claimed in claim 1 , wherein detecting, by the electronic device, the at least one event to transfer the private content into the secure folder of the electronic device comprises:
detecting, by the electronic device, that the displayed content is the private content; receiving, by the electronic device, at least one input on the screen of the electronic device; determining, by the electronic device, whether the at least one input is authorized; performing, by the electronic device, at least one of: initiating transfer of the displayed content from the non-secure folder to the secure folder of the electronic device in response to determining that the at least one input is authorized; and not initiating transfer of the displayed content from the non-secure folder to the secure folder of the electronic device in response to determining that the at least one input is not authorized.

7 . The method as claimed in claim 1 , wherein the at least one event on the electronic device comprises at least one gesture of the user of the electronic device to shift the private content into the secure folder of the electronic device, and at least one gesture of the user of the electronic device to switch quickly between the non-secure folder of the electronic device and the secure folder of the electronic device.

8 . The method as claimed in claim 7 , wherein switching from the non-secure folder of the electronic device to the secure folder of the electronic device requires a user authentication, and switching from the secure folder of the electronic device to the non-secure folder of the electronic device does not require the user authentication.

9 . The method as claimed in claim 1 , wherein the method further comprises: automatically recommending at least one suitable action based on the displayed content being detected as the private content.

10 . A

Assignees

• Samsung Electronics Co Ltd

Inventors

• MOHARANA Sukumar
• Goyal Manoj
• Mohanty Debi Prasanna
• Laipubam Herojit Sharma
• Ramena Gopi
• Thota Siva Prasad
• Vala Vanraj
• Anand Chinmay
• Panwar Archit
• Rao Vipin
• Purre Naresh
• Nalam Ravi Chandra
• Bhamidipati Sreevatsa Dwaraka
• Reddy Kachana Raghunatha

Classifications

• G06F21/31

  User authentication · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06F16/54

  Browsing; Visualisation therefor · CPC title