The project establishes a junior research group to investigate new and improved methods of information processing for automated understanding of human behavior. Human behavior includes all externally perceivable activities such as postures, gestures and facial expressions that are shown consciously or unconsciously. From the behavior we also want to draw conclusions about underlying inner affective state of a person.

Field of the project HUBA.

 Publikationen

• Philipp Werner, Frerk Saxen, Ayoub Al-Hamadi
  Facial Action Unit Recognition in the Wild with Multi-Task CNN Self-Training for the EmotioNet Challenge
  IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2020. We almost won the EmotionNet 2020 challenge! Video
• Philipp Werner, Frerk Saxen, Ayoub Al-Hamadi, Hui Yu
  Generalizing to Unseen Head Poses in Facial Expression Recognition and Action Unit Intensity Estimation
  International Conference on Automatic Face and Gesture Recognition (FG), 2019. Acceptance rate: 40%
• Frerk Saxen, Sebastian Handrich, Philipp Werner, Ehsan Othman, Ayoub Al-Hamadi
  Detecting Arbitrarily Rotated Faces for Face Analysis
  International Conference on Image Processing (ICIP), 2019.

• Ehsan Othman, Frerk Saxen, Dmitri Bershadskyy, Philipp Werner, Ayoub Al-Hamadi, Joachim Weimann
  Predicting Group Contribution Behaviour in a Public Goods Game from Face-to-Face Communication
  Sensors, June 2019. Impact-Factor (2018): 3.031
• Sebastian Handrich, Philipp Waxweiler, Philipp Werner, Ayoub Al-Hamadi
  3D Human Pose Estimation Using Stochastic Optimization in Real Time
  International Conference on Image Processing (ICIP), 2018.

Contact: 

 Philipp Werner, Ayoub Al-Hamadi