Tri-axial Camera Array for 3D Face Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face recognition and iris recognition systems are suboptimal for 2D face recognition and 3D eye location, prone to performance reduction due to ambient lighting, and are often bulky, with a need for devices that can perform effectively in various lighting conditions and minimize size while maintaining low calculation times and operating costs.
Innovation Solution
A device comprising three cameras with their optical axes arranged such that the first and second cameras form an angle less than 10°, and the third camera's optical axis intersects the first and second, allowing for 2D face recognition and 3D face position determination through triangulation, while avoiding upward camera orientations to mitigate lighting issues and minimize size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cameras are arranged with coinciding optical axes to maximize triangulation volume, then 3D reconstruction capability is improved, but device size increases and lighting performance deteriorates
Solution Approach 1:
The patent transitions from a single-plane camera arrangement to a three-dimensional spatial configuration where cameras are offset along an axis perpendicular to the image plane. This dimensional change enables triangulation functionality while maintaining a compact footprint, as the cameras utilize the depth dimension rather than requiring lateral expansion.
Solution Approach 2:
The patent employs electronic image processing and software-based triangulation algorithms to dynamically compute 3D position information from images captured by cameras in a compact offset arrangement. This dynamic computational approach replaces static geometric requirements, allowing accurate 3D reconstruction without requiring cameras to be physically positioned at widely separated locations.
2Measurement precision
If cameras are arranged with coinciding optical axes to maximize triangulation volume, then 3D reconstruction capability is improved, but device size increases
Solution Approach 1:
The patent transitions from a single-plane camera arrangement to a three-dimensional spatial configuration where cameras are offset along an axis perpendicular to the image plane. This dimensional change enables triangulation functionality while maintaining a compact footprint, as the cameras utilize the depth dimension rather than requiring lateral expansion.
Solution Approach 2:
The patent employs electronic image processing and software-based triangulation algorithms to dynamically compute 3D position information from images captured by cameras in a compact offset arrangement. This dynamic computational approach replaces static geometric requirements, allowing accurate 3D reconstruction without requiring cameras to be physically positioned at widely separated locations.
3Adaptability or versatility
If camera optical axis points upwards to maximize acquisition volume, then field of view is improved, but performance deteriorates due to ambient lighting
Solution Approach 1:
The patent applies different orientation strategies to different cameras based on their specific functions. Face recognition cameras are positioned to optimize for frontal face capture with controlled lighting, while the third camera is oriented to provide triangulation capability. Each camera's orientation is locally optimized for its specific role rather than all cameras sharing a uniform upward orientation.
Solution Approach 2:
The patent divides the camera system into functionally distinct groups: cameras dedicated to face recognition with optimized lighting conditions, and a third camera dedicated to triangulation. This segmentation allows each subsystem to be optimized independently, with face recognition cameras avoiding upward orientations that would expose them to ambient lighting interference.
4Adaptability or versatility
If multiple separate systems are used for face recognition and iris recognition, then functional versatility is improved, but device complexity and size increase
Solution Approach 1:
The patent implements a unified camera system that performs multiple biometric functions. The same camera array used for face recognition also captures images for iris recognition and provides triangulation data for 3D positioning. This multi-functional approach eliminates the need for separate dedicated systems, reducing overall device complexity while maintaining comprehensive biometric capabilities.
Solution Approach 2:
The patent merges face recognition, iris recognition, and 3D triangulation functions into a single integrated camera system. By combining these previously separate functions into one unified hardware platform, the system reduces complexity, minimizes the number of components, and simplifies data processing pipelines while maintaining all required biometric capabilities.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The device achieves robust performance in both 2D face recognition and 3D face position determination, is less affected by ambient lighting, and is compact, reducing calculation times and operating costs by using a common set of cameras for both tasks.
Implementation Method 1
Such systems enable triangulation to be performed and thus make it possible to perform three-dimensional (3D) reconstruction of the position of a zone of the subject's face by using cameras having fields of view that overlap.
Data Source
AI summary
A device for use in identifying or authenticating a subject positioned in an acquisition volume on the basis of at least one biometric characteristic of the subject, the device including in succession, in offset manner along a camera placement axis: a first camera; a second camera; and a third camera; the optical axes of the first and second cameras forming between them an angle strictly less than 10°, and the optical axis of the third camera intersecting the optical axes of the first and second cameras, the optical axes of the first and second cameras each forming an angle less than or equal to 5° relative to a normal axis perpendicular to the camera placement axis.