Audiovisual Document Boundary Detection and Memory Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audiovisual recording systems, such as PVRs, often record unwanted content due to timing inaccuracies, leading to inefficient use of memory space and user inconvenience in identifying the start and end of desired documents.

Innovation Solution

A method that allows users to select key images to identify the start and end of an audiovisual document, with a probability value associated with each sequence shot to guide the user in choosing the correct markers, and subsequent erasure of non-document content to free up memory space.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If time margins are added to ensure complete document recording, then document completeness is improved, but memory space efficiency deteriorates

Engineering Contradiction:
Improvedocument completenessVSAvoidmemory space efficiency
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The patent extracts and removes the unwanted margin content (foreign content before document start and after document end) from the recorded sequence. By identifying characteristic images that mark the actual beginning and end of documents, the system separates desired content from unwanted margin content, deleting only the latter to free memory space while preserving complete documents.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary analysis during the recording phase by detecting characteristic images and their positions. This advance identification of document boundaries allows the system to prepare for efficient later processing, knowing exactly where margin content begins and ends without requiring manual user intervention.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If manual positioning of start and end marks is used to delimit documents, then recording precision is improved, but user time consumption deteriorates

Engineering Contradiction:
Improvedocument delimitation precisionVSAvoiduser time consumption
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs automatic document delimitation by autonomously detecting characteristic images and calculating their positions within the recorded sequence. Instead of requiring manual user positioning, the system serves itself by automatically identifying document boundaries and preparing selection menus, significantly reducing user time consumption while maintaining high precision.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical process of user scrolling and positioning with an automated image recognition and calculation system. The system uses characteristic image detection and mathematical calculations to automatically determine document boundaries, substituting human manual operations with automated computational processes.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If automatic detection of specific sequences is used, then user operation ease is improved, but document identification accuracy deteriorates

Engineering Contradiction:
Improveautomatic delimitation easeVSAvoiddocument identification accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The system provides feedback to the user by displaying a selection menu showing multiple detected characteristic images with their positions and probabilities of being document boundaries. This feedback mechanism allows users to review automated detections and make informed selections, combining automatic detection efficiency with user verification for high accuracy.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Instead of relying on a single automatic detection, the system performs excessive detection by identifying multiple characteristic images and presenting them as options. This partial action approach, where multiple potential boundaries are detected and displayed, allows the system to exceed basic automatic detection and provide users with choices, improving accuracy while maintaining ease of operation.

Inventive Principle:
Principle #16Partial or excessive action

4Measurement precision

If probability values are displayed for each sequence shot to guide user selection, then selection accuracy is improved, but interface complexity deteriorates

Engineering Contradiction:
Improvemarker selection accuracyVSAvoidinterface complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies local quality by providing detailed probability information specifically at the points where users need decision-making support (the characteristic image selection menu), rather than making the entire interface complex. The probability values are localized to the relevant selection context, adding precision where needed without overwhelming the overall user interface.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP1958202B1Method for identifying a document recorded by a display, selection of key images and an associated receptor
Publication Date: 2012.09.26 THOMSON LICENSING SA
  • EP1958202B1 patent drawingFigure 1
  • EP1958202B1 patent drawingFigure 2~2.6
  • EP1958202B1 patent drawingFigure 3

AI summary

The invention relates to a method for identifying an audio-visual document consisting in programming an audio-visual content recording by a user in a recording device for recording a determined document, the recording being over, in detecting and displaying by the device identifiers assigned to sequence shots extracted from the recorded content, wherein each sequence shot exhibits at least one determined characteristic, in displaying a probability indication associated to each sequence shot for assisting a user in identifying the beginning or the end of said document, in introducing by the user an instruction for selecting a displayed identifier and in identifying the beginning or the end of the document by the sequence shot associated to the selected identifier. A receptor provided with a user interface for carrying out said method is also disclosed.