Intelligent Saliency-based Deep Pedestrian Tracking System for Advanced Driving Assistance

In an attempt to make the new generations of cars ever safer, scientific investigation is recently focused on monitoring the driver's attention level strongly correlated to the cardiac activity regulated by Autonomic Nervous System (ANS) of the subject. In this field, the Photoplethysmography (PPG) has proved to be an excellent non-invasive detector of the subject's psychological state. Specifically, we are interested on the development of intelligent driver attention monitor applied over the driving scenarios involving pedestrians. The identification and tracking of pedestrians in the driving scene is a widely investigated problem in the scientific community with particularly efficient solutions although often computationally expensive. In this context, the authors propose a full intelligent pipeline for the identification, monitoring and tracking of the salient pedestrians combined with a physio-based attention monitoring system. The collected experiment results confirmed the effectiveness of the proposed pipeline.