Generative AI Text Generation from Screen-Change OCR

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video conference systems struggle with sharing and retrieving text-based content effectively, making it difficult to check or summarize meeting content later, especially for security purposes, as only compressed video is shared, and there is no efficient method to automatically generate meeting minutes or identify security-related information.

Innovation Solution

A method and apparatus using generative AI to automatically generate meeting minutes, action items, and questions/answers by detecting screen changes, extracting text from shared screens using OCR, combining text frames, and integrating speech data to create topic-based summaries and Q&A.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If screen sharing is performed using video compression (e.g., H.264), then content sharing flexibility is improved, but text content retrieval and search capability deteriorates

Engineering Contradiction:
Improvecontent sharing flexibilityVSAvoidtext content retrieval capability
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent extracts text information from the compressed video frames through OCR (optical character recognition) technology. By extracting text from individual frames and assembling them into a searchable text document, the system retrieves text content without requiring video playback, thus resolving the contradiction between maintaining video compression benefits and enabling text retrieval.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary processing layer between video compression and text retrieval. This intermediary system captures frames from the compressed video, processes them through OCR, and generates a separate text-based searchable document. This intermediary step enables text retrieval without compromising the original video compression efficiency.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If compressed video is shared during meetings, then sharing efficiency is improved, but automatic document generation and meeting minutes creation deteriorates

Engineering Contradiction:
Improvesharing efficiencyVSAvoidautomatic document generation capability
Core Design Contradiction:
ProductivityVSExtent of automation

Solution Approach 1:

The patent performs preliminary actions by capturing video frames during the meeting and extracting text in real-time or near-real-time. This preliminary processing enables automatic document generation after the meeting without requiring manual transcription or summary creation, thus improving automation extent while maintaining sharing efficiency.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system enables self-service by automatically generating meeting minutes and documents using the extracted text and integrated AI capabilities. The system serves itself by processing the shared video content and producing structured meeting documentation without human intervention, thereby improving automation while preserving the efficiency benefits of compressed video sharing.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If video frames are processed to extract text, then text extraction accuracy is improved, but processing time and computational resources increase

Engineering Contradiction:
Improvetext extraction accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the video into individual frames and processes only the relevant frames containing text content. By segmenting rather than processing the entire video continuously, and by using frame-based OCR only where needed, the system achieves high text extraction accuracy while reducing overall processing time and computational resource consumption.

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enhances the accuracy of generating meeting minutes and allows effective monitoring of shared data, enabling security officers to identify meeting content and manage page changes, improving the retrieval and security of shared information.

Implementation Method 1

extracting text from a changed frame

Methodology Applied
Scientific EffectOptical Character Recognition:

Data Source

PatentEP4642022A1Method, apparatus, and computer program for automatically generating text using generative ai
Publication Date: 2025.10.29 SAMSUNG SDS CO LTD
  • EP4642022A1 patent drawingFigure 1
  • EP4642022A1 patent drawingFigure 2
  • EP4642022A1 patent drawingFigure 3

AI summary

The disclosure relates to a method, an apparatus, and a computer program for automatically generating text using generative AI. A method for automatically generating text using generative AI of the disclosure may include: determining whether or not there is a change between frames constituting images; extracting text from a changed frame; and generating first text by combining the extracted text with previously extracted text.