Image Object Segmentation from Click Input for Faster Editing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual selection of objects in digital images is cumbersome and time-consuming, especially when editing large numbers of images or those with many objects, due to the difficulty in accurately defining the outline of the object's area.
Innovation Solution
A computer-implemented method that uses object detection and segmentation modules to automatically identify primary and secondary object regions in an image, allowing users to select and edit objects through user input, with the system processing the image to identify precise object regions using machine learning models like YOLO-V6 COCO and segmentation models like Efficient-Vit-SAM.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual selection process is used to define object outlines, then users can select specific objects for editing, but the process becomes cumbersome and time-consuming
Solution Approach 1:
The system performs automatic object detection and segmentation without requiring manual user input for outlining. The processing units automatically identify primary object regions and secondary object regions based on user selections, allowing the system to serve itself in the object selection task rather than requiring continuous user guidance.
Solution Approach 2:
The system pre-processes the image to identify and segment multiple object regions before the user makes their final selection. By performing object detection and segmentation in advance, the system prepares potential object candidates, so when the user selects a position, the appropriate object is already identified and ready for editing.
2Measurement precision
If manual outlining is used to precisely define object areas, then accurate object selection is achieved, but the process becomes difficult to carry out accurately
Solution Approach 1:
The patent replaces the mechanical manual outlining process with automated computer vision algorithms. Instead of requiring users to manually trace object boundaries with cursor operations, the system uses machine learning models (object detection and segmentation) to automatically define precise object boundaries, substituting computational processing for manual mechanical interaction.
Solution Approach 2:
The system introduces an intermediary processing layer between the user's simple position selection and the final object identification. The object detection and segmentation modules act as intermediaries that translate a rough user position selection into a precise object boundary definition, mediating between imprecise user input and precise object selection.
3Productivity
If automatic object detection is implemented, then editing speed is improved, but the system complexity increases with multiple processing modules
Solution Approach 1:
The system segments the object identification task into two distinct modules: object detection for identifying primary object regions, and object segmentation for identifying secondary object regions within those primary regions. This segmentation of functionality allows each module to specialize in a specific aspect of object identification, improving overall accuracy and speed while organizing complexity into manageable, independent components.
Data Source
AI summary
Described herein is a computer implemented method including displaying an image on a display and then processing, using one or more processing units, the image to identify one or more primary object regions in the image. The method further includes receiving a first user input selecting a first input image position, determining that the first input image position does not correspond to any primary object region, and in response to determining that the first input image position does not correspond to any primary object region, processing the image based on the first input image position to identify a secondary object region in the image.