Suggestive Marking for Object Selection in Image Editing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image editing software requires precise marking techniques to select objects, which can be cumbersome and error-prone, especially on compact devices like laptops and mobile terminals with smaller screens, leading to frustration and inefficiency.
Innovation Solution
A method and apparatus for identifying a desired object in an image using suggestive marking, where a casual annotation or gesture suggests the object's boundary, allowing for the generation and display of multiple output images for user selection, utilizing edge detection algorithms to refine the object identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional precise marking techniques are used to select objects, then object selection accuracy is improved, but operation complexity and time consumption increase
Solution Approach 1:
The system performs preliminary object detection and generates candidate object lists before the user makes a selection. By pre-processing the image to identify potential objects and their boundaries, the system reduces the user's workload from performing precise manual marking to simply selecting from pre-generated candidates, thereby significantly reducing time consumption while maintaining selection accuracy
Solution Approach 2:
The patent introduces an intermediary computational layer between the user's simple marking gesture and the final object selection. This intermediary layer processes the rough user marking, detects edges, identifies candidate objects, and presents refined options to the user, thereby mediating between low-precision user input and high-precision object selection requirements
2Measurement precision
If conventional precise marking techniques are used to select objects, then object selection accuracy is improved, but ease of operation deteriorates
Solution Approach 1:
The system enables self-service by automatically performing object detection, boundary identification, and candidate generation based on the user's simple marking. The computational system serves itself by processing the rough user input and automatically refining it into precise object selections, eliminating the need for the user to manually perform complex precision marking tasks
Solution Approach 2:
An intermediary processing layer automatically transforms the user's simple marking gesture into precise object boundary identification. This intermediary layer handles the complex edge detection and object recognition tasks, allowing the user to operate with simple gestures while the system performs the sophisticated processing needed for accurate object selection
3Productivity
If conventional marking techniques are used on compact devices, then no significant improvement is achieved, but user frustration and errors increase
Solution Approach 1:
The patent replaces the mechanical precision-marking system with a computational image processing system. Instead of relying on the user's manual dexterity to precisely trace object boundaries on small screens, the system uses automated edge detection algorithms and computer vision techniques to identify objects, thereby substituting mechanical precision requirements with computational processing capabilities that work effectively on compact devices
Solution Approach 2:
An intermediary computational layer processes the user's simple marking input and automatically refines it into accurate object selections. This intermediary layer compensates for the limitations of compact device screens and input methods by performing sophisticated image processing that enhances both productivity and selection reliability on mobile and tablet devices
Data Source
AI summary
A computer implemented method and apparatus for identifying a desired object of an image by using a suggestive marking. The method comprises receiving a first marking to an image, the first marking suggesting a desired object of the image and the desired object being defined by a boundary; generating, based on the first marking, a plurality of output images, wherein each image of the plurality of output images indicates a computer-identified object; and displaying the plurality of output images.