Eye-Gaze Differences and Saliency Prediction Across Different age Groups

摘要

The human visual system has developed the ability to process a scene by selecting the most relevant parts of the scene unconsciously. This mechanism is called selective attention, and has been developed in order to quickly spot danger. Only within the last 10-15 years, researchers in computer vision exploited the concept of human selective attention for computational models of saliency prediction. However, the saliency models developed so far have considered only adult observers and do not generalize well to the observers of different age groups. We developed a framework to analyze eye-movement behavior of observers belonging to different age groups while they explored the scene. Analysis results were used to develop age adapted saliency models. Our results suggest that the proposed age-adapted saliency model outperforms existing saliency models in predicting the regions of interest across age groups.