Automated Presentation Slide Generation for Mobile Devices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing presentation software requires substantial user input to create visually appealing presentations, which is impractical on mobile devices with limited input options and resource constraints, necessitating a method to automatically determine and prioritize important images for efficient layout arrangement.

Innovation Solution

The solution involves analyzing pixel properties, such as luminance, to quickly differentiate between functional and non-functional images, applying specific layout rules to enhance visibility of functional images, including scaling, positioning, and z-order adjustments, without relying on complex image recognition algorithms, thereby minimizing resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If complex image recognition algorithms are used to identify functional images, then the accuracy of image classification improves, but the resource consumption and processing time increase significantly

Engineering Contradiction:
Improveimage classification accuracyVSAvoiddevice resource consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent segments the image analysis process into two distinct stages: a fast preliminary filtering stage using simple pixel property analysis (luminance, color saturation) to identify candidate functional images, and a more detailed analysis stage only for those candidates. This segmentation allows the system to achieve accurate classification while minimizing overall resource consumption by applying complex algorithms only when necessary.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the analysis parameters from complex semantic image recognition to simple physical pixel properties (luminance values, color saturation). By transforming the classification criteria from high-level image understanding to low-level pixel statistics, the system achieves functional image identification with minimal computational resources while maintaining sufficient accuracy for presentation layout purposes.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If extensive user input is required to create visually appealing presentations, then the quality and customization of presentations improve, but the ease of operation deteriorates

Engineering Contradiction:
Improvepresentation qualityVSAvoiduser input requirement
Core Design Contradiction:
Manufacturing precisionVSEase of operation

Solution Approach 1:

The patent implements self-service by enabling the presentation software to automatically analyze uploaded images, classify them as functional or non-functional based on pixel properties, and apply appropriate layout rules without requiring user intervention. The system serves itself by making intelligent decisions about image placement, scaling, and positioning, thereby maintaining high presentation quality while minimizing user input requirements.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent incorporates feedback mechanisms where the system continuously monitors image properties and layout effectiveness, automatically adjusting the arrangement of elements based on the classified importance of each image. This closed-loop approach allows the software to self-optimize the presentation quality without extensive user guidance, adapting to different image types and contexts automatically.

Inventive Principle:
Principle #23Feedback

3Manufacturing precision

If intelligent decisions are made to optimize presentation design, then the visual appeal of presentations improves, but the processing time and computational load increase

Engineering Contradiction:
Improvedesign qualityVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-classifying images into functional and non-functional categories using simple pixel property analysis before the actual layout arrangement process. This preliminary classification enables subsequent layout operations to proceed efficiently with predetermined rules, avoiding the need for complex real-time decision-making during the layout generation phase and thereby reducing overall processing time.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2709067B1Method for automatically generating presentation slides containing picture elements
Publication Date: 2019.05.01 BLACKBERRY LTD
  • EP2709067B1 patent drawingFigure 1
  • EP2709067B1 patent drawingFigure 2~3
  • EP2709067B1 patent drawingFigure 4~5

AI summary

A computer-implemented method for intelligently generating presentations, the method comprising the steps of: analysing the variation in colours of pixels in the digital image; and, if the variation is below a predetermined threshold, applying one or more rules to increase visibility of the digital image in the presentation.