Facial Motion Detection via Importance-Based Feature Point Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing facial motion detection technologies face reduced accuracy due to the use of uniformly corrected images that do not account for varying importance of face regions during different facial motions, leading to suppression of important motion detection and decreased recognition accuracy.
Innovation Solution
The method involves specifying preferable feature points for each facial motion based on importance information and the magnitude of motion, generating corrected images for each motion, and inputting these to a recognition model for improved detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If uniformly corrected images are used for facial motion detection, then the correction process is simple and fast, but the detection accuracy decreases due to suppression of important motion regions
Solution Approach 1:
The patent applies local quality by differentiating the correction process across different facial regions. Importance information is generated to identify key facial regions (eyes, eyebrows, mouth) that require preserved motion characteristics. The correction process is then applied selectively - regions with high importance values maintain their original motion patterns while less important regions undergo standard correction, thereby preserving critical facial expressions while maintaining overall image correction.
2Area of stationary object
If all feature points are used for image correction, then the correction covers the entire face, but important motion information is suppressed due to uniform treatment of all regions
Solution Approach 1:
The system generates importance information that assigns different weights to different facial regions. Feature points are then selected based on these importance values - regions with high importance (such as eyes and mouth during specific expressions) are preserved with minimal correction, while other regions undergo full correction. This selective approach ensures comprehensive coverage while maintaining reliability of critical expression detection.
Solution Approach 2:
The facial image is segmented into multiple regions based on importance information, with different correction strategies applied to each segment. Key regions identified through importance mapping receive specialized handling to preserve motion characteristics, while other regions undergo standard correction procedures. This segmentation allows the system to treat different parts of the face differently according to their diagnostic value for expression recognition.
Data Source
AI summary
An image processing method executed by a computer, the method includes detecting a plurality of feature points of a face from an input image, referring to importance information that indicates an importance of a region within an image in a process of detecting a predetermined facial motion from the image, selecting, from the plurality of feature points detected by the detecting, one or more points that correspond to an image region including an importance indicated by the importance information equal to or smaller than a first threshold value, correcting the input image by using the one or more points selected by the selecting, to generate a corrected image; and determining whether or not the predetermined facial motion is occurring in the input image, based on an output obtained by inputting the corrected image to a recognition model.