Smartphone Image Retention via PIQUE Quality Scoring and User Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Smartphones face challenges in processing large numbers of images captured in burst mode, with many images being of poor quality due to noise and blurring, and existing automated best shot selection methods do not adequately align with user perception of image quality.
Innovation Solution
A system and method that utilizes a Perception based Image Quality Evaluator (PIQUE) module to determine a no-reference quality score for images based on block weights, presents this score and a spatial quality map to users for feedback, and learns user perception to retain or delete images within predefined quality thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated best shot selection is used to process burst images, then images of good quality can be identified based on objective parameters, but the selection may not align with user perception of image quality
Solution Approach 1:
The system implements a feedback mechanism where users provide feedback on the quality of images selected by the automated system. This feedback is then used to train a machine learning model that learns user preferences and adjusts the quality assessment algorithm accordingly, aligning automated selection with user perception over time
Solution Approach 2:
The system transitions from using fixed objective parameters for quality assessment to a dynamic parameter set that incorporates learned user preferences. The quality metrics are adjusted based on feedback data, allowing the system to adapt to individual user perceptions of image quality
2Extent of automation
If computational methods are used to process and separate good quality images from poor quality images, then automated selection can be achieved, but the computation becomes complex especially when a large number of images need to be processed
Solution Approach 1:
The system performs preliminary quality assessment using fast objective metrics before presenting images to users for feedback. This preliminary filtering reduces the computational burden by quickly eliminating obviously poor quality images, requiring detailed processing only for borderline cases
Solution Approach 2:
The machine learning model continuously learns from user feedback and improves its own performance automatically. The system refines its quality assessment algorithm through self-training on collected feedback data, reducing the need for complex manual tuning and ongoing computational resources
Data Source
AI summary
An automated system and method for retaining images in a smart phone are disclosed. The system may then determine a no-reference quality score of the image using a PIQUE module. The PIQUE module utilizes block level features of the image to determine the no-reference quality score. The system may present the image and the no-reference quality score to the user and accept a feedback towards quality of the image. The system may utilize a supervised learning model for continually learning a user's perception of quality of the image, the no-reference quality score determined by the PIQUE module, and the user feedback. Based on the learning, the supervised learning model may adapt the no-reference quality score and successively the image may either be retained or isolated for deletion, based on the adapted quality score and a predefined threshold range.