3D Face Model Matching for Orientation Variance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face authentication systems struggle with matching images where the face is not facing forward, as the orientation of preregistered face images does not align with captured images, leading to difficulties in feature recognition and reduced matching accuracy.
Innovation Solution
A matching support apparatus and method that utilizes a reference face development image generated from three-dimensional data to display and designate facial features, allowing for feature information generation and matching across varying orientations, enabling accurate feature recognition and candidate selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If forward-facing face images are used for registration, then matching accuracy is improved, but the system cannot handle images with different face orientations
Solution Approach 1:
The patent transforms the two-dimensional face image into a three-dimensional face model by detecting feature points and constructing spatial coordinates. This dimensional transformation allows the system to represent facial features in 3D space, enabling accurate matching regardless of the face's orientation in the captured image. The 3D coordinates of feature points remain consistent even when the face rotates, solving the orientation adaptability problem while maintaining matching accuracy.
Solution Approach 2:
The patent changes the representation parameters from 2D image coordinates to 3D spatial coordinates. By expressing facial feature positions in three-dimensional space (x, y, z coordinates) rather than two-dimensional image planes, the system can accommodate various face orientations. The 3D parameter representation allows the matching algorithm to compare facial features based on their spatial relationships, which remain invariant under rotation, thus resolving the contradiction between accuracy and adaptability.
2Measurement precision
If discrete feature sites are used for authentication, then authentication accuracy is improved, but feature position changes with face orientation making matching difficult
Solution Approach 1:
The patent detects discrete facial features (moles, scars, wrinkles) and represents their positions in three-dimensional space rather than on a two-dimensional image plane. By establishing 3D coordinates for each feature point based on the face's geometric structure, the system captures the spatial relationships that remain stable across different orientations. This dimensional transformation ensures that feature positions are defined by their intrinsic spatial coordinates, not by their projected positions on a 2D image, thereby maintaining feature position stability while preserving authentication accuracy.
Solution Approach 2:
The patent performs preliminary construction of a three-dimensional face model and pre-establishes the spatial coordinates of discrete feature sites before matching occurs. By preparing the 3D geometric framework in advance and calculating the spatial positions of features relative to this framework, the system creates a stable reference system that is independent of the captured image's orientation. This preliminary action ensures that feature positions are determined by their inherent spatial relationships rather than by the variable orientation of the captured face, thus maintaining stability for accurate authentication.
Data Source
AI summary
A matching support apparatus 10 has a generation unit 11 that, in the case where a feature region indicating a facial feature of a person targeted for matching visible on the skin surface of the person targeted for matching is designated, using a reference face development image display region, displayed on the screen of a display device 30, for displaying a reference face development image generated based on three-dimensional data of a head serving as a reference, generates feature information relating to the designated feature region, a matching unit 12 that matches the feature information against matching information in which a matching-use face development image and matching-use feature region for each person registered in advance are associated with each other, and a selection unit 13 for selecting a person to serve as a candidate, based on a matching result.