Machine Vision based Stress, Cognition, and Emotion Sensing in Driver/Occupant Monitoring Systems
The management and transfer of vehicular control between a driver and in-built computational intelligence is a key challenge as the automotive roadmap evolves fully autonomous vehicles. Today we can determine some basic attributes of a driver – analysis of the eye-region enables a determination of drowsiness and gaze. However, the next generation of driven monitoring systems need an improves understanding of the humans within a vehicle – their emotional and cognitive states, their interactions, both with the vehicle itself and between one another, and the levels of driver stress. Today, machine vision can only provide rudimentary solutions, but a fusion of machine vision systems with bio-signals (EEG, heart rate & breathing) and multi-modal imaging, such as thermal vision, can provide a basis for new data-driven machine vision systems that can more accurately and reliably understand the humans in a vehicular environment. This session seeks to explore some of the emerging potential of data-driven machine vision systems in the context of the next generation of non-contact, in-cabin driver monitoring technologies.
Chairman
Prof. Peter Corcoran (IEEE Fellow), C3I
Imaging Centre, College of Science & Engineering, National
University of Ireland Galway (www.nuigalway.ie/c3i)
Prof. Corcoran is an IEEE Fellow, recognized for his contributions
to digital camera technology and is recognized as the #1 Computer
Science researcher in Ireland by www.research.com
Dr. Joseph Lemley, Director of Research, Xperi Corporation, Ireland
(www.xperi.com)
Dr. Lemley leads a research cluster of c.20 engineers at Xperi
corporation (www.Xperi.com) with a focus on advanced Driver &
Occupant monitoring technologies.
Prof. Arcangelo Merla, Dipartimento di Ingegneria e Geologia,
University of Chieti-Pescara (www.unich.it/)
Prof. Merla is one of the leading researchers in Europe on thermal
vision systems and their application in parasympathetic behavioural
analysis.
The sessions covers (but is not limited to)
papers on
–
Machine-vision algorithms, neural models and techniques for improved
driver sensing
– Multi-modal imaging for driver and occupant behaviours
(NIR, thermal imaging, neuromorphic [event camera] imaging systems &
techniques, including hybrid imaging)
– Data Annotation, Augmentation and Acquistion techniques for MV
based driver sensing
– New Datasets for MV based sensing of cognitive, emotional or
stress of subjects
– Training methodologies or experimental frameworks for neural
machine vision algorithms and models
Important Dates
Notification October 10, 2022