Intelligent Autofocus Subject Detection via Verbal Input

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Intelligent autofocus technologies face challenges in accurately identifying and shifting focus between subjects in digital images, especially when using self-timers, requiring manual user intervention and struggling to differentiate between multiple subjects.

Innovation Solution

A detection apparatus and method that utilizes an input and output interface to receive verbal inputs for subject identification, processes these inputs to determine attributes such as name, color, shape, and position, and applies corresponding detection methods to automatically focus on the desired subject in digital images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If intelligent AF automatically identifies and focuses on a subject from captured images, then the operation simplicity is improved, but the ability to shift focus to a different subject deteriorates

Engineering Contradiction:
Improveoperation simplicityVSAvoidfocus shifting ability
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system dynamically switches between automatic focus selection (when no touch is detected) and manual focus selection via touch input (when touch is detected). This allows the focus subject to be automatically determined in normal operation while enabling user intervention when focus shifting is needed, resolving the contradiction between operational simplicity and focus shifting capability

Inventive Principle:
Principle #15Dynamics

2Extent of automation

If intelligent AF focuses on a subject from multiple subjects in a captured image, then the automation extent is improved, but the difficulty of detecting and measuring the correct subject deteriorates

Engineering Contradiction:
Improveautomatic focus selectionVSAvoidsubject identification accuracy
Core Design Contradiction:
Extent of automationVSDifficulty of detecting and measuring

Solution Approach 1:

The system provides visual feedback by displaying a focus selection screen that shows multiple detected subjects with their identification information. The user can see which subjects are detected and can select the desired focus subject by touching the corresponding area, ensuring accurate subject identification while maintaining automation

Inventive Principle:
Principle #23Feedback

3Ease of operation

If self-timer is activated for image capture, then the ease of operation is improved, but the ability to shift focus after activation deteriorates

Engineering Contradiction:
Improveself-timer operationVSAvoidfocus adjustment capability
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system allows the user to pre-select the focus subject by touching the desired area on the focus selection screen before the self-timer captures the image. This preliminary focus selection ensures that even though the self-timer is running, the correct subject will be in focus when the image is captured, maintaining both ease of operation and focus adjustment capability

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10027878B2Detection of object in digital image
Publication Date: 2018.07.17 EARTH SAMA INC
  • US10027878B2 patent drawing
  • US10027878B2 patent drawing
  • US10027878B2 patent drawing

AI summary

A detection apparatus for detecting a subject in a digital image, including: an input and output interface to which a word for identifying the subject is input; and a processor configured to: receive the word for identifying the subject from the input and output interface; determine an attribute that is specified by the word for identifying the subject, the attribute being at least one of name, color, shape, size, and position; select a method of detection in accordance with the determined attribute; apply the selected method of detection to the digital image to detect the subject in the digital image; and output a result of the detection of the subject so as to cause a prescribed action to be performed with respect to the detected subject in the digital image.