Three high-impact publications have brightened the somewhat gloomy COVID-19 days at our department

“Classical and Deep Learning Methods for Recognizing Human Activities and Modes of Transportation with Smartphone Sensors “, published in Information Fusion (IF 10.7), describes our approach to activity recognition that brought us victory at the Sussex-Huawei Locomotion Recognition Challenge, an international competition in activity recognition.

“Machine Learning and End-to-end Deep Learning for Monitoring Driver Distractions from Physiological and Visual Signals” (https://ieeexplore.ieee.org/document/9062481), published in IEEE Access (IF 4.1) studies how to detect distracted driving – a highly relevant topic in the time of transitioning from fully human to autonomous driving, when driving aids often contribute to drivers being less focused.

“Towards Cognitive Load Inference for Attention Management in Ubiquitous Systems” (https://ieeexplore.ieee.org/document/9067018), published in IEEE Pervasive Computing (IF 3.8), presents two methods for detecting cognitive load with unobtrusive sensors, and explores how such methods could be used for compter-aided attention management.