Panoramic Pre-Attentive Cameras with Mirror-Directed Face Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social robots face challenges in seamlessly interacting with humans due to the trade-off between wide-field visual awareness and high spatial acuity for facial recognition, especially in the far field, which existing approaches struggle to address effectively.
Innovation Solution
A system utilizing panoramic pre-attentive sensing with wide-angle cameras and attentive sensing with a high-resolution, narrow-field camera and a mirror-based gaze deflection system to capture high-resolution images, allowing for efficient face recognition in both near and far fields.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If wide-angle cameras are used for panoramic sensing, then wide-field visual awareness is improved, but spatial acuity for facial recognition deteriorates
Solution Approach 1:
The visual system is segmented into two functional components: a wide-field pre-attentive camera for panoramic awareness and a narrow-field attentive camera for high-resolution facial recognition. This segmentation allows each component to specialize in its optimal function without compromise.
Solution Approach 2:
A mirror-based gaze deflection system acts as an intermediary, redirecting the attentive camera's field-of-view to track and focus on persons of interest detected by the pre-attentive camera. This intermediary enables the narrow-field camera to access distant targets without requiring a wide inherent field-of-view.
2Measurement precision
If narrow-field high-resolution cameras are used for facial recognition, then spatial acuity is improved, but wide-field visual awareness deteriorates
Solution Approach 1:
The system dynamically switches between pre-attentive and attentive modes based on detected persons of interest. The attentive camera's field-of-view is dynamically redirected using the mirror assembly to track moving targets, allowing high-resolution imaging of specific persons while maintaining awareness of the broader scene through the pre-attentive camera.
3Adaptability or versatility
If panoramic imaging is used for wide coverage, then detection of persons in various directions is improved, but recognition accuracy in the far field deteriorates
Solution Approach 1:
The mirror assembly serves as an optical intermediary that enables the attentive camera to access far-field persons of interest detected by the pre-attentive camera. By redirecting light paths, the mirror allows high-resolution imaging of distant targets without requiring the attentive camera to have a wide inherent field-of-view.
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 system achieves effective panoramic face recognition up to 35 meters, providing high-resolution imaging and improved social awareness for mobile robots, enhancing their ability to detect and recognize individuals in various environments.
Implementation Method 1
a mirror mounted on a controllable structure to direct a gaze of the mirror to a specified area of the panoramic view
Data Source
AI summary
There is provided a method, system, and device for capturing high resolution images of a scene for analysis of one or more persons. The method includes: receiving one or more pre-attentive images, the one or more pre-attentive images capturing the scene; detecting one or more persons in the one or more pre-attentive images; determining a feature vector and a geo-location for at least one of the detected persons in the pre-attentive image; matching the feature vector and geo-location to a previously detected person for tracking of such detected person, and where there is no match, initializing a tracking of a new person; receiving an attentive image that captures the detected person by directing gaze at a specified azimuthal location, the attentive image comprising a smaller field-of-view (FoV) than the one or more pre-attentive images; and outputting the attentive image.