IP Library Granted Patent US 12,182,980
Granted Patent B2
US 12,182,980 · App. 17/187,934 · Granted Dec 31, 2024

Cognitive filtering of content from shared screen display

Inventors: Thangadurai Muthusamy (Bangalore, IN); Chourasia Abhishek Kumar (Bangalore, IN); Saravanan Devendran (Bangalore, IN); Pietro Iannucci (Rome, IT)
Assignee: Kyndryl, Inc.
G06T7/0002G06F3/1454G06F18/2178G06F18/24G06F18/40G06N5/04G06N20/00G06T5/20G06T5/70G06T5/77G06T2207/10016G06T2207/20081G06V30/10
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 12,182,980
App. No.
17/187,934
Granted
Dec 31, 2024
Kind
B2
Abstract

Methods, computer program products, and/or systems are provided that perform the following operations: obtaining source display content; identifying one or more objectionable portions included in the source display content; modifying the one or more objectionable portions included in the source display content to filter content in the one or more objectionable portions; generating a filtered display content, wherein the filtered display content includes the one or more modified objectionable portions and one or more non-objectionable portions of the source display content; and providing the filtered display content for rendering of a shared presentation.

Claims (78)

1. A computer-implemented method comprising:

obtaining source display content, the source display content including shared screen content rendered by and being displayed on a screen of a source device and being output by the source device for concurrent display thereof by a remote device located remotely from the source device;

identifying one or more objectionable portions included in the source display content;

modifying the one or more objectionable portions included in the source display content to filter content in the one or more objectionable portions;

generating a filtered display content, wherein the filtered display content includes the one or more modified objectionable portions and one or more non-objectionable portions of the source display content; and

providing the filtered display content for rendering of filtered shared screen content by the remote device while the shared screen content is being displayed on the screen of the source device in original form.

2. The computer-implemented method of claim 1 , wherein identifying the one or more objectionable portions included in the source display content comprises:

obtaining text data and image data in the source display content; and

classifying some portions of the source display content as potential objectionable content and classifying other portions of the source display content as potential non-objectionable content based, at least in part, on the text data and image data.

3. The computer-implemented method of claim 2 , further comprising:

providing the text data and the image data to an image classifier model, wherein the image classifier model is trained to classify the source display content as objectionable content, as non-objectional content, or some of the content as objectionable content and some of the content as non-objectionable content;

identifying portions of the source display content as objectionable; and

providing the identified portions of the source display content to an image reconstruction model, wherein the image reconstruction model is trained to filter the one or more objectionable portions to obscure content and generate the filtered display content.

4. The computer-implemented method of claim 3 , further comprising:

providing one or more of the objectionable portions to the image reconstruction model;

filtering the one or more objectionable portions to generate one or more modified objectionable portions including obscured content;

providing the one or more non-objectionable portions to the image reconstruction model; and

generating the filtered display content based on the one or more non-objectionable portions and the one or more modified objectionable portions.

5. The computer-implemented method of claim 4 , further comprising:

providing preview filtered display content based on the filtered display content;

obtaining indications that one or more portions of the preview filtered display content were misclassified; and

providing data associated with the one or more misclassified portions for use in updating the image classifier model.

6. The computer-implemented method of claim 1 , wherein modifying the one or more objectionable portions included in the source display content comprises:

masking the one or more objectionable portions to generate the one or more modified objectionable portions, wherein the masking obscures content in the one or more objectionable portions.

7. The computer-implemented method of claim 1 , further comprising:

obtaining feedback data associated with the filtered display content, the feedback data including one or more indications of objectionable content misclassified as non-objectionable; and

updating an image classifier model and an image reconstruction model based, at least in part, on the feedback data.

8. The computer-implemented method of claim 7 , wherein the feedback data includes user preference data.

9. The computer-implemented method of claim 7 , wherein the feedback data includes historical shared screen content interaction data associated with the filtered display content.

10. The computer-implemented method of claim 1 , wherein the shared screen content rendered by and displayed on the screen of the source device includes a presentation controlled by a user at the source device during a video conference.

11. A computer program product comprising a computer readable storage medium having stored thereon:

program instructions programmed to obtain source display content, the source display content including shared screen content being displayed on a screen of a source device and being output for concurrent display thereof by a remote device located remotely from the source device;

program instructions programmed to identify one or more objectionable portions included in the source display content;

program instructions programmed to modify the one or more objectionable portions included in the source display content to filter content in the one or more objectionable portions;

program instructions programmed to generate a filtered display content, wherein the filtered display content includes the one or more modified objectionable portions and one or more non-objectionable portions of the source display content;

program instructions programmed to provide the filtered display content for rendering of filtered shared screen display content by the remote device while the shared screen content is being displayed on the screen of the source device in original form;

program instructions programmed to obtain feedback data associated with the filtered display content, the feedback data including one or more indications of non-objectionable content misclassified as objectionable; and

program instructions programmed to update an image classifier model and an image reconstruction model based, at least in part, on the feedback data.

12. The computer program product of claim 11 , wherein identifying the one or more objectionable portions included in the source display content comprises:

obtaining text data and image data in the source display content; and

classifying some portions of the source display content as potential objectionable content and classifying other portions of the source display content as potential non-objectionable content based, at least in part, on the text data and image data.

13. The computer program product of claim 12 , the computer readable storage medium having further stored thereon:

program instructions programmed to provide the text data and the image data to the image classifier model, wherein the image classifier model is trained to classify the source display content as objectionable content, as non-objectional content, or some of the content as objectionable content and some of the content as non-objectionable content;

program instructions programmed to identify portions of the source display content as objectionable; and

program instructions programmed to provide the identified portions of the source display content to the image reconstruction model, wherein the image reconstruction model is trained to filter the one or more objectionable portions to obscure content and generate the filtered display content.

14. The computer program product of claim 13 , the computer readable storage medium having further stored thereon:

program instructions programmed to provide one or more of the objectionable portions to the image reconstruction model;

program instructions programmed to filter the one or more objectionable portions to generate one or more modified objectionable portions including obscured content;

program instructions programmed to provide the one or more non-objectionable portions to the image reconstruction model; and

program instructions programmed to generate the filtered display content based on the one or more non-objectionable portions and the one or more modified objectionable portions.

15. The computer program product of claim 11 , wherein modifying the one or more objectionable portions included in the source display content comprises:

masking the one or more objectionable portions to generate the one or more modified objectionable portions, wherein the masking obscures content in the one or more objectionable portions.

16. The computer program product of claim 10 , the computer readable storage medium having further stored thereon:

program instructions programmed to provide preview filtered display content based on the filtered display content;

program instructions programmed to obtain indications that one or more portions of the preview filtered display content were misclassified; and

program instructions programmed to provide data associated with the one or more misclassified portions for use in updating the image classifier model.

17. A computer system comprising:

a processor set; and

a computer readable storage medium;

wherein:

the processor set is structured, located, connected and programmed to run program instructions stored on the computer readable storage medium; and

the stored program instructions include:

program instructions programmed to obtain source display content, the source display content including shared screen content being displayed on a screen of a source device and being output for concurrent display thereof by a remote device located remotely from the source device;

program instructions programmed to identify one or more objectionable portions included in the source display content;

program instructions programmed to modify the one or more objectionable portions included in the source display content to filter content in the one or more objectionable portions;

program instructions programmed to generate a filtered display content, wherein the filtered display content includes the one or more modified objectionable portions and one or more non-objectionable portions of the source display content;

program instructions programmed to provide the filtered display content for rendering of filtered shared screen content by the remote device while the shared screen content is being displayed on the screen of the source device;

program instructions to obtain feedback data associated with the filtered display content, the feedback data including one or more indications of non-objectionable content misclassified as objectionable; and

program instructions to update an image classifier model and an image reconstruction model based, at least in part, on the feedback data.

18. The computer system of claim 17 , wherein identifying the one or more objectionable portions included in the source display content comprises:

obtaining text data and image data in the source display content; and

classifying some portions of the source display content as potential objectionable content and classifying other portions of the source display content as potential non-objectionable content based, at least in part, on the text data and image data.

19. The computer system of claim 17 , wherein modifying the one or more objectionable portions included in the source display content comprises:

masking the one or more objectionable portions to generate the one or more modified objectionable portions, wherein the masking obscures content in the one or more objectionable portions.

20. The computer system of claim 18 , wherein the stored program instructions further include:

program instructions programmed to provide the text data and the image data to the image classifier model, wherein the image classifier model is trained to classify content as objectionable content, as non-objectional content, or some of the content as objectionable content and some of the content as non-objectionable content;

program instructions programmed to identify portions of the source display content as objectionable; and

program instructions programmed to provide the identified portions of the source display content to the image reconstruction model, wherein the image reconstruction model is trained to filter the one or more objectionable portions to obscure content and generate the filtered display content.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 18, 2021
From: INTERNATIONAL BUSINESS MACHINES CORPORATION
To: KYNDRYL, INC.
Reel/Frame 058213/0912 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 1, 2021
From: MUTHUSAMY, THANGADURAI; KUMAR, CHOURASIA ABHISHEK; DEVENDRAN, SARAVANAN; IANNUCCI, PIETRO
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 055440/0205 →