Face Detection in Live Viewfinder for Satisfactory Image Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Capturing a digital image with multiple persons in the scene where all facial expressions are satisfactory is challenging due to the difficulty in coordinating smiles and open eyes, especially as the number of individuals increases.
Innovation Solution
An apparatus and method that detect faces in captured images, display them on a live viewfinder feed, and automatically select candidate faces for replacement, allowing for real-time evaluation and selection of satisfactory facial expressions without user input, enabling the storage of a digital still image with optimal facial expressions from multiple captured images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple still images are captured to increase the probability of satisfactory facial expressions, then the quality of the final image improves, but the time required to capture a satisfying image increases
Solution Approach 1:
The system performs preliminary actions by capturing multiple still images in rapid succession before the user needs to select the best one. These pre-captured images are then used to generate candidate faces, allowing the system to prepare multiple options in advance rather than requiring sequential capture and evaluation.
Solution Approach 2:
The system creates copies of facial features from different captured images to construct composite candidate faces. By copying and combining facial elements (eyes, smile, cheeks) from multiple sources, the system can generate idealized candidate faces that represent the best possible expression without requiring the perfect moment to occur naturally.
2Ease of operation
If the system provides detailed face detection and selection options to improve user control, then the ease of selecting satisfactory faces improves, but the device complexity increases
Solution Approach 1:
The system performs self-service by automatically detecting faces, generating candidate faces, and even auto-selecting the best candidate without requiring extensive user intervention. The automatic face detection and candidate generation processes handle the complex tasks independently, reducing the operational burden on the user while maintaining system sophistication.
Solution Approach 2:
The system segments the face selection process into distinct functional modules: face detection, candidate face generation, candidate presentation, and selection. This segmentation allows each module to be optimized independently and managed separately, reducing overall system complexity while providing comprehensive functionality.
3Reliability
If the system displays multiple candidate faces from captured images, then the quality of the final selected image improves, but the difficulty of detecting and measuring satisfactory expressions increases
Solution Approach 1:
The system introduces an intermediary processing stage that transforms raw captured images into standardized candidate faces with consistent formatting and features. This intermediary step normalizes the data from multiple sources, making it easier to detect and compare facial expressions across different images without dealing with the raw complexity of individual captures.
Solution Approach 2:
The system changes parameters by transforming captured images into candidate faces with standardized characteristics. By adjusting and normalizing facial features (alignment, scaling, lighting) during the candidate generation process, the system simplifies the detection and measurement tasks while maintaining the quality information from multiple original captures.
Data Source
AI summary
According to one aspect, there is provided an apparatus comprising at least one processing unit and at least one memory. The at least one memory stores program instructions that, when executed by the at least one processing unit, cause the apparatus to obtain a set of captured still images captured with a camera, detect at least one face in the captured still images, and cause display of the detected at least one face on a display providing a live viewfinder feed of a scene to be captured as an additional display element while providing the live viewfinder feed on the display.