Eye Tracking for Remote Presentation Engagement

The patent introduces an intermediary eye tracking system that acts as a mediator between the presenter and the remote audience. The system captures eye movement data from attendees and translates it into actionable engagement metrics, allowing presenters to detect disengagement without direct visual observation. This intermediary layer resolves the contradiction by providing precise measurement through specialized hardware while keeping the core presentation system unchanged.

The patent replaces the mechanical/visual system of direct presenter observation with an optical-electronic eye tracking system. Instead of the presenter visually scanning the room to assess attention, the system uses cameras and image processing to automatically track eye positions and calculate engagement metrics. This substitution provides more precise and continuous measurement while reducing the manual effort required.

The patent extracts only the essential features needed for engagement detection from the full eye tracking data stream. Instead of processing all raw eye movement information, the system focuses on calculating divergence values between control and non-control parties and determining disengagement values based on specific metrics. This selective extraction reduces computational requirements while maintaining the ability to provide real-time feedback.

The patent implements partial processing by focusing on key engagement indicators rather than analyzing every aspect of eye movement data. The system calculates divergence and disengagement values as representative metrics that capture the essential engagement state without requiring exhaustive analysis of all available data, thus reducing computational load while maintaining feedback effectiveness.

The patent segments the audience into control and non-control parties, allowing for separate analysis of their eye movement patterns. By dividing the audience group and calculating divergence between these segments, the system achieves more accurate disengagement detection. The segmentation approach breaks down the complex measurement task into manageable comparisons between defined groups, improving precision without overwhelming complexity.

The patent applies different analysis approaches to different audience segments (control vs. non-control parties). Instead of treating all attendees uniformly, the system calculates divergence specifically between these groups to identify disengagement patterns. This localized quality approach allows for more accurate measurement by considering the specific characteristics and expectations of different audience segments.