1 code implementation • 8 Nov 2023 • Thorsten Hempel, Magnus Jung, Ahmed A. Abdelrahman, Ayoub Al-Hamadi
We tackle this challenge and present NITEC, a hand-annotated eye contact dataset for ego-vision interaction.
1 code implementation • 14 Sep 2023 • Thorsten Hempel, Ahmed A. Abdelrahman, Ayoub Al-Hamadi
Together with new accumulated training data that provides full head pose rotation data and a geodesic loss approach for stable learning, we design an advanced model that is able to predict an extended range of head orientations.
Ranked #1 on Head Pose Estimation on CMU Panoptic + 300W-LP
no code implementations • 13 Jul 2023 • Laslo Dinges, Marc-André Fiedler, Ayoub Al-Hamadi, Thorsten Hempel, Ahmed Abdelrahman, Joachim Weimann, Dmitri Bershadskyy
Due to limited training data, we also utilize discriminative models for deception detection.
no code implementations • 10 Jan 2023 • Thorsten Hempel, Laslo Dinges, Ayoub Al-Hamadi
Emotion expressions serve as important communicative signals and are crucial cues in intuitive interactions between humans.
no code implementations • 7 Nov 2022 • Thorsten Hempel, Marc-André Fiedler, Aly Khalifa, Ayoub Al-Hamadi, Laslo Dinges
Current technological advances open up new opportunities for bringing human-machine interaction to a new level of human-centered cooperation.
no code implementations • 8 Mar 2022 • Thorsten Hempel, Ayoub Al-Hamadi
This way, we are able generate additional meaningful information that allows to achieve higher-level tasks, while simultaneously leveraging the view-invariance of object detections to improve the accuracy and the robustness of the odometry estimation.
1 code implementation • 7 Mar 2022 • Ahmed A. Abdelrahman, Thorsten Hempel, Aly Khalifa, Ayoub Al-Hamadi
In this paper, we propose a robust CNN-based model for predicting gaze in unconstrained settings.
Ranked #2 on Gaze Estimation on Gaze360 (using extra training data)
1 code implementation • 25 Feb 2022 • Thorsten Hempel, Ahmed A. Abdelrahman, Ayoub Al-Hamadi
In this paper, we present a method for unconstrained end-to-end head pose estimation.
Ranked #2 on Head Pose Estimation on BIWI