Application of Emotion Recognition Methods in Automotive Research

1. Application of Emotion Recognition Methodsin Automotive Research Matthias Wimmer, Christian Peter, J�rg Voskamp, Martin A. Tischler Psychologist, Working on my dissertation; Research interest: Driving pleasure � factors that determine/affect emotional state of the drivers Not an expert for programming algorythms, I present you today one of our studies, we did in spring this year. Overview: - Some general words about driving pleasure - Methods that can be used in vehicles - Study about emotion recognition pleasure of driving a car is very important for success of a vehicle model Measuring emotional reactions of drivers in close-to-reality environments opens extended views of the interaction driver-vehicle information to design vehicles to better satisfy constantly increasing customer demands my subject area is positive emotionsPsychologist, Working on my dissertation; Research interest: Driving pleasure � factors that determine/affect emotional state of the drivers Not an expert for programming algorythms, I present you today one of our studies, we did in spring this year. Overview: - Some general words about driving pleasure - Methods that can be used in vehicles - Study about emotion recognition pleasure of driving a car is very important for success of a vehicle model Measuring emotional reactions of drivers in close-to-reality environments opens extended views of the interaction driver-vehicle information to design vehicles to better satisfy constantly increasing customer demands my subject area is positive emotions

2. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 2 Theoretical Background of Driving Pleasure Receive general rules to design a car. Models tries to explain the development of positive emotions Structure of emotions that come of the interaction with products. Model what wants to describe how positive emotions develope drivers attempt to maintain a level of task difficulty It should be noticed that there is almost no research done on the positive aspects of driving a car.Receive general rules to design a car. Models tries to explain the development of positive emotions Structure of emotions that come of the interaction with products. Model what wants to describe how positive emotions develope drivers attempt to maintain a level of task difficulty It should be noticed that there is almost no research done on the positive aspects of driving a car.

3. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 3 Measurement of Emotions in Vehicles Change a setup of the car. Need for easy to use methods, Emotion manifests on different levels.Change a setup of the car. Need for easy to use methods, Emotion manifests on different levels.

4. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 4 Pilot Study �Driving Pleasure W204� The aim of the study was to evaluate technologies to quantify driving pleasure in a close-to-reality scenario. common studies only use retrospective questionnaires and interviews product development purposes new methods to quantify the emotional state of the driver are needed.The aim of the study was to evaluate technologies to quantify driving pleasure in a close-to-reality scenario. common studies only use retrospective questionnaires and interviews product development purposes new methods to quantify the emotional state of the driver are needed.

5. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 5 Cooperation The objective of this pilot study was to apply sophisticated emotion recognition techniques into a test car and to demonstrate their applicability for the described purpose. Three research institutes cooperated and provided two cars with sensory equipment to conduct a study in a close-to-reality environment. For safety reasons it was decided to do the study on a separated proving ground in Southern Germany.The objective of this pilot study was to apply sophisticated emotion recognition techniques into a test car and to demonstrate their applicability for the described purpose. Three research institutes cooperated and provided two cars with sensory equipment to conduct a study in a close-to-reality environment. For safety reasons it was decided to do the study on a separated proving ground in Southern Germany.

6. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 6 Subjects: 8 non professional drivers (age: 33-53) 4 married couples Test track was too expensive4 married couples Test track was too expensive

7. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 7 Task:Drive on three courses We booked a proving ground in southern germany for a weekendWe booked a proving ground in southern germany for a weekend

8. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 8 Technical Setup

9. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 9 The Methods in Detail a wireless and easy to use sensor system for collecting the emotion-related physiological parameters skin resistance, skin temperature, and heart rate; EREC system (for Emotion RECognition) used in several studies of different application fields, inside and outside a lab, and proved to be stable and reliable. Both cars were equipped with cell phones and hands free sets, in order to give instructions to the drivers For logging the actual position and the basic driving dynamic data the GPS-based Racelogic VBOX Mini was used.a wireless and easy to use sensor system for collecting the emotion-related physiological parameters skin resistance, skin temperature, and heart rate; EREC system (for Emotion RECognition) used in several studies of different application fields, inside and outside a lab, and proved to be stable and reliable. Both cars were equipped with cell phones and hands free sets, in order to give instructions to the drivers For logging the actual position and the basic driving dynamic data the GPS-based Racelogic VBOX Mini was used.

10. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 10 Subjective Driving Experience Fulfills my inputsFulfills my inputs

11. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 11 Facial Expression Recognition Recognizing 6 emotions; Stimme 2 Dimensionen determine the six universal emotions specified by Ekman and to give evidence about the magnitude of the visible expression An Active Shape Model represents the contour of the face and it is fit to the camera image via learned objective functions. Successfully tracking a human face during some period of time, our system determines the motion of the individual parts such as eyes, chin, cheeks, and lips. that drivers don't express this emotion strongly while being alone in the car. The presence of non emotional-related head motion while observing the traffic makes facial expression recognition very difficult.Recognizing 6 emotions; Stimme 2 Dimensionen determine the six universal emotions specified by Ekman and to give evidence about the magnitude of the visible expression An Active Shape Model represents the contour of the face and it is fit to the camera image via learned objective functions. Successfully tracking a human face during some period of time, our system determines the motion of the individual parts such as eyes, chin, cheeks, and lips. that drivers don't express this emotion strongly while being alone in the car. The presence of non emotional-related head motion while observing the traffic makes facial expression recognition very difficult.

12. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 12 Speech Analysis Follw another approach � 2 dimesional model of emotion For analysing the data, techniques from data mining and knowledge discovery have been applied. The extracted information then is used to build two classifiers based on the valence/arousal model, one for detecting valence and one for arousal. They were trained on 10% of the collected data and tested using 10-fold cross-validation. They were then applied on the remaining 90%.Follw another approach � 2 dimesional model of emotion For analysing the data, techniques from data mining and knowledge discovery have been applied. The extracted information then is used to build two classifiers based on the valence/arousal model, one for detecting valence and one for arousal. They were trained on 10% of the collected data and tested using 10-fold cross-validation. They were then applied on the remaining 90%.

13. Tischler, Peter, Wimmer & Voskamp, Application of Emotion Recognition Methods in Automotive Research, 2nd Workshop "Emotion and Computing" 13 Conclusions Methods show corresponding results Speech analysis Easy to apply, but speech is not a natural product while driving Facial expression recognition Challenge in fitting the face model and classifying the facial expression Problems: changing light and background, non emotional-related head movements, driver is talking Advantages of tested methods Non disturbing Continuous data collection Useful completion to psychological methods brightnessbrightness