Image Vividness Scoring to Remove Distracting Color Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image modification techniques fail to effectively identify and reduce vivid, distracting colors in digital images that are not associated with the subject, leading to a distraction from the main subject of the photograph.
Innovation Solution
A method involving vividness scoring, agglomeration of candidate pixels, and subject determination to modify or remove vivid regions that do not overlap the image subject, using machine-learned models and image processing techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If vividness scoring is performed on all pixels to identify distracting regions, then the ability to reduce distracting colors is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent divides the image processing task into distinct stages: first performing vividness scoring on all pixels to identify candidate pixels, then agglomerating these candidates into regions, and finally determining which regions to modify based on subject overlap. This segmentation allows the system to manage computational complexity by breaking down the overall task into manageable sub-tasks that can be processed sequentially
Solution Approach 2:
The patent extracts only the pixels that meet the vividness threshold criteria into a separate candidate set, rather than processing all pixels equally throughout the entire pipeline. This extraction step reduces the data volume for subsequent agglomeration and subject comparison operations, thereby managing computational load while maintaining detection accuracy
2Object-affected harmful factors
If all vivid regions are removed to eliminate distractions, then the distraction from subject is reduced, but the subject itself may be altered if vivid regions overlap with it
Solution Approach 1:
The patent implements a feedback mechanism where the identified vivid regions are compared against the detected subject regions, and the modification decision is adjusted based on this comparison. If a vivid region overlaps with the subject, the system feedbacks this information to prevent modification of that region, thereby protecting the subject integrity while still removing non-overlapping distracting regions
Solution Approach 2:
The patent applies different treatment to different regions of the image based on their relationship to the subject. Regions that are vivid but do not overlap with the subject are modified to reduce distraction, while regions that are vivid and do overlap with the subject are preserved. This local differentiation allows the system to reduce harmful distractions while maintaining subject integrity
3Measurement precision
If manual editing is used to remove vivid regions, then precision in preserving subject is improved, but productivity and time efficiency deteriorate
Solution Approach 1:
The patent implements an automated system that performs vividness scoring, region agglomeration, subject detection, and modification decision-making without requiring manual intervention. The system serves itself by automatically identifying distracting regions and applying appropriate modifications while preserving the subject, thereby achieving both high productivity through automation and high precision through sophisticated algorithms
Solution Approach 2:
The patent uses vividness scoring that evaluates multiple parameters (color saturation, brightness, hue) to automatically identify distracting regions. By changing and analyzing these color parameters programmatically, the system achieves precision comparable to manual editing while maintaining high processing speed and productivity
Data Source
AI summary
Methods and systems for modifying a digital image are described herein. The method can include performing vividness scoring for a plurality of pixels of the digital image, determining one or more candidate pixels based on the vividness scoring for the plurality of pixels, and agglomerating the one or more candidate pixels into one or more suggested agglomerates. The method can also include determining at least one subject of the digital image, removing at least one agglomerate from the one or more suggested agglomerates based on at least one of the at least one subject of the digital image or one or more characteristics of the at least one agglomerate, generating a modified digital image with the one or more suggested agglomerates modified, and outputting the modified digital image.