Eye Gaze Feedback for Remote Meeting Engagement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video conferencing systems lack the ability to effectively track participant attention and provide real-time feedback to presenters, leading to inefficient engagement and adaptation of content during remote meetings, especially in unconstrained environments with varying head poses and illumination.
Innovation Solution
A remote meeting system that utilizes consumer-level webcams to capture images, extracts facial landmarks and eye information, and employs convolutional neural networks to determine eye gaze angles, providing feedback to presenters on regions of interest within the presentation content, allowing for personalized interaction and content adaptation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If consumer-level webcams are used to capture participant images in unconstrained environments, then the system becomes more accessible and easier to deploy, but the precision of eye gaze angle determination deteriorates due to varying head poses and illumination conditions
Solution Approach 1:
The system performs preliminary detection of facial landmarks and eye positions from webcam images before calculating gaze angles. This preliminary processing establishes a baseline reference frame that compensates for varying head poses and illumination conditions, enabling accurate gaze determination despite environmental constraints
Solution Approach 2:
The system dynamically adjusts processing parameters based on detected conditions, including head pose estimation and illumination analysis. By changing parameters such as feature detection thresholds and coordinate transformation matrices according to environmental conditions, the system maintains measurement precision across varying constraints
2Productivity
If real-time eye gaze tracking is implemented during remote meetings, then presenter feedback and content adaptation are improved, but the computational complexity and processing requirements increase
Solution Approach 1:
The system extracts only the critical features necessary for gaze determination, such as facial landmarks and eye positions, from the full webcam image stream. By taking out and processing only these essential elements rather than analyzing entire images, the system reduces computational complexity while maintaining real-time feedback capability
Solution Approach 2:
The system performs partial processing by focusing computational resources on key regions of interest (eyes and facial features) rather than processing the entire image. This selective partial action provides sufficient information for gaze tracking without the excessive computational burden of full-image analysis
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of providing feedback to a presenter in a remote meeting includes capturing images of remote participants in the remote meeting using cameras associated with computing devices that display content presented by the presenter. Eye gaze angle information for at least one of the remote participants is determined based on the captured images. At least one region of interest in the displayed content is identified based on the eye gaze angle information. Feedback is provided to the presenter including an indication of the identified at least one region of interest.