Screen Wakeup Using Neural Network Face Gaze Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing screen wakeup technologies face challenges in achieving accurate screen activation without relying on high-quality pupil images, which are often affected by low-quality images from front-facing cameras, thereby increasing production costs.
Innovation Solution
A screen wakeup method utilizing a preconfigured neural network to analyze multiple image frames and determine whether the face image matches a preset face and belongs to a user gazing at the screen, thereby switching the screen from an off to an on state.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a general front-facing camera is used to obtain pupil image for identification, then device cost is reduced, but identification accuracy deteriorates due to low image quality
Solution Approach 1:
The patent changes the identification parameters from pupil-based (requiring high-quality images) to face-based (working with lower-quality images). By using face image features instead of pupil features, the system can achieve accurate identification with standard front-facing cameras, resolving the contradiction between cost and accuracy
Solution Approach 2:
The patent uses face image as a substitute (copy) for pupil image. Instead of requiring the original high-quality pupil image, the system creates an alternative identification pathway using face image data, which can be obtained from standard cameras without compromising identification functionality
2Measurement precision
If a special iris camera is used to obtain high-quality pupil image, then identification accuracy is improved, but device production cost increases
Solution Approach 1:
The patent extracts the identification function from the pupil image domain and relocates it to the face image domain. By separating the identification task from the requirement of high-quality pupil imaging, the system eliminates the need for expensive specialized cameras while maintaining identification accuracy
Solution Approach 2:
The patent makes the front-facing camera universal by enabling it to perform both its original function (capturing face images for display/wake-up) and the additional function of providing identification data. This multi-functionality eliminates the need for separate specialized iris cameras, reducing device cost
3Adaptability or versatility
If pupil image-based identification is used, then gaze detection capability is achieved, but image quality requirement increases device complexity
Solution Approach 1:
The patent merges the wake-up function and identification function into a single integrated process using face images. By combining these functions and using the same face image data for both purposes, the system achieves gaze detection capability without requiring separate specialized hardware, thereby reducing device complexity
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
This application provides a screen wakeup method and apparatus. The screen wakeup method includes: obtaining M image frames, where each image frame includes a first face image, and M is an integer greater than or equal to 1; determining, based on a preconfigured neural network, whether each first face image matches a preset face image and belongs to a user who is gazing at a screen of a device; and when each first face image matches the preset face image and belongs to the user, switching the screen from a screen-off state to a screen-on state. According to the technical solutions provided in this application, accuracy of screen wakeup of a device can be improved without significantly increasing costs.