Snapshots of the future: Tool learns to predict user's gaze in headcam footage

(Institute of Industrial Science, The University of Tokyo) University of Tokyo researchers have developed a computational tool that can learn from headcam footage of complex tasks to predict where the user's future gaze will be focused. This tool combines 'visual-saliency' mapping of frames of footage based on distinctive visual features with 'gaze-prediction' mapping based on head movement and previous gaze direction. This tool could facilitate real-time guidance derived from headcam footage in situations involving complex tasks such as surgery and manufacturing.
Source: EurekAlert! - Social and Behavioral Science - Category: International Medicine & Public Health Source Type: news