MA Zwischenvortrag, Shashank Rao, Ort: O27/5202, Datum: 02.10.2018, Zeit: 10:30 Uhr

The intent behind this thesis is to do a qualitative analysis of current infrastructure and techniques for emotion detection and create models that combine the discussed modalities for emotion detection (e.g., driver inattention, facial expressions, physiological signals and driving contexts) in order to improve the accuracy in detection of negative emotions and states (frustration/anger/disgust, panic/fear and boredom/fatigue/inattentiveness). The models obtained provide us with further information that can improve the quality of predictions as emotion detection still remains as a very fuzzy and a difficult task. This will facilitate our aim to improve driver safety for future cars through its interfaces.