Dual-Shot Image Preview Using Algorithmic Face Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in capturing high-quality dual-shot images due to challenges in setting proper shot compositions, especially with the front image being lopsided and the small size of the front image PIP making it hard to confirm quality, leading to poor composite image quality.
Innovation Solution
An electronic device apparatus that utilizes algorithmic recognition of a subject portion, such as a user's face, to set an area of pixels of interest within the front image, outputting a partial image for overlay on the rear image, facilitating real-time preview and composition of dual-shot images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the front image is displayed in a PIP box over the rear image, then the dual-shot composition is achieved, but the front image becomes lopsided and difficult to compose properly
Solution Approach 1:
The system performs preliminary action by displaying a preview of the front image separately before final composition, allowing users to adjust framing and composition in advance. The preview screen shows the front image with guiding markers indicating the optimal PIP box position, enabling users to compose the front image correctly before it is overlaid on the rear image.
2Shape
If the front image PIP box is made small to fit the composite image, then the dual-shot format is maintained, but the user cannot confirm the quality of the sub-image
Solution Approach 1:
The system resolves this contradiction by adding another dimension to the display - showing the front image in a separate full-screen preview mode before composition. This allows users to examine the sub-image quality in detail at full resolution, then proceed to compose the final dual-shot image with the appropriately sized PIP box.
3Ease of operation
If the user manually triggers capture with one hand while steadying the device with the other, then the capture function is operational, but the device sways causing out-of-focus images
Solution Approach 1:
The system applies self-service by enabling automatic capture functionality that triggers based on detected conditions (such as face detection, gesture recognition, or timer-based triggering). This eliminates the need for manual button pressing during capture, allowing the device to remain steady and focused while the capture is automatically initiated based on pre-set conditions or user-prepared states.
4Ease of operation
If algorithmic recognition is used to identify the subject portion, then the composition is improved, but computational requirements and power consumption increase
Solution Approach 1:
The system applies partial action by implementing algorithmic recognition selectively - using simplified algorithms for basic functions (such as face detection for preview positioning) and reserving more computationally intensive algorithms for specific scenarios. The preview function uses lightweight algorithms that consume minimal power, while full composition algorithms are only activated when needed for final image processing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A dual-shot image preview method and apparatus is disclosed. An electronic device with dual image sensors provides a preview screen combining in real-time a partial image captured from a first image sensor and a rear image captured from a second image sensor. The partial image is a portion of a front image that includes a desired portion of the user, such as the user's face based on algorithmic detection. The electronic device displays a dual-shot preview screen by overlaying the partial image on the rear image.