This topic is formulated based on the data analysis phase in SmartHelm research project. From the data acquisition phase conducted as part of the summer study with SmartHelm experimental set-up various data streams are collected such as eye-tarcking data (eye-gaze information), speech signals, etc ., raw eye-tracking information consists eye gaze measurement values in time series format for the complete duration of the experiment. Similarly raw speed data contains uttering values i.e at which time instance did rider gave a speech input using the Automatic speech recognition (ASR). In addition, location information (GPS) data, eye-tracking video data are also availabe which specifies the route in which the study is carried out. In total, the study is conducted with 20  participants and the complete study all together contains approximately 40 datasets. In principle, eye tracking facilitates understanding behavior aspects in which direction does the bicycle rider focusing during the ride. In-depth research in this field aims to research most important questions such as what is the Area of Interest (AOI) and how frequently is the rider focussing on the route?

AIM: Processing, preparing and analyzing the multimodal data to visualize the behavior of the rider on map.

The factors influencing the Field of Vision should be visualized with eye gaze information, as an example one of the state of art technique applied in this field are the parallel Scan path technique[Ref. 3]. At the beginning, step literature study should be carried to study what are the various state-of-the-art techniques available for visualizing eye-tracking data. Then with available other relevant data GPS data and eye-tracking video data, the obtained results from the eye gaze information should be intgrated with GPS and video data to visualize in different kinds of the visualization forms such as Heat maps, graphical categorization, speech map rtc., The main steps in the thesis can be specified as follows:

• Study of the eye-tracking data

• Literature study to find the various techniques

• Selection of the significant methods based on dependency

• Visualization in different forms

• Determining the KPI‘s

Integrating GPS and video data for visualization
 

References

1. Berger, Martin, and Linda Dörrzapf. "Sensing comfort in bicycling in addition to travel data." Transportation research procedia 32 (2018): 524-534.

2. Trefzger, Mathias, Tanja Blascheck, Michael Raschke, Sarah Hausmann, and  Thomas Schlegel. "A visual comparison of gaze behavior from pedestrians  and cyclists." In Proceedings of the 2018 ACM symposium on eye tracking research & applications, pp. 1-5. 2018.

3. Michael Raschke, Dominik Herr, Tanja Blascheck, Michael Burch, Michael Schrauf, Sven Willmann, and Thomas Ertl. 2014. A Visual Approach for Scan Path Comparison. In Proceedings of the Symposium on Eye Tracking Research & Applications. ACM, 135–142.

• Informatik Bereich
• Praktisch Erfahrung mit Multimodal Daten
• Motiviert
• Deutsch oder Englisch beides möglich