Face Database Management for Restricted Memory Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital image pickup devices, such as camera phones, face memory limitations when trying to store diverse facial images for effective face recognition, leading to performance degradation due to varying poses, expressions, and lighting conditions.
Innovation Solution
A method and apparatus for managing a reference face database by selecting and registering new facial images based on classification conditions, such as pose, expression, and lighting, and deleting less effective previously stored images to optimize memory usage and improve recognition rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If diverse facial images are pre-registered in the database to improve face recognition performance, then recognition rate is improved, but memory resources are exhausted
Solution Approach 1:
The patent applies parameter changes by evaluating facial images based on multiple parameters including pose angles, lighting conditions, and facial expressions. The system selectively registers images that satisfy specific parameter thresholds, optimizing the balance between recognition rate and memory usage by only storing images that provide meaningful diversity for recognition improvement
Solution Approach 2:
The patent implements local quality by categorizing facial images into different groups based on their specific characteristics (front view, side view, lighting conditions, expressions). Instead of uniformly storing all images, the system evaluates and stores only those images that provide local quality improvements for specific recognition scenarios, thereby optimizing memory utilization
2Quantity of substance
If facial images are selectively registered based on classification conditions to optimize memory usage, then memory efficiency is improved, but recognition performance may degrade
Solution Approach 1:
The patent applies preliminary action by pre-defining classification conditions and evaluation criteria for facial images before the actual registration process. The system establishes threshold values for pose angles, lighting conditions, and expression variations in advance, enabling automated selective registration that maintains recognition performance while optimizing memory efficiency
Solution Approach 2:
The patent implements feedback mechanisms by evaluating the contribution of each newly registered facial image to the overall recognition performance. The system monitors recognition rates and adjusts the selection criteria dynamically, ensuring that only images that provide measurable improvement are registered, thereby maintaining high recognition performance with efficient memory usage
Data Source
AI summary
Various embodiments of the present disclosure presents a method for managing a reference face database to improve face recognition performance under a restricted memory environment, said method comprising: acquiring a new input face image; determining a classification condition corresponding to the input face image with reference to classification conditions of existing registered face images of a reference face database used for face recognition; and if the classification condition corresponding to the input face image is a specific classification condition and the value of the face image corresponding to the specific classification condition in the reference face database is less than a preset threshold value, selecting the input face image as an image that can be additionally registered to the reference face database.