Arian Beckmann is a research associate at the "Computer Vision & Graphics" Group at Fraunhofer Heinrich Hertz Institute (HHI). He graduated in Mathematics at the University of Muenster with a Master of Science in 2021. His research interests include generative modelling, adversarial machine learning, probabilistic machine learning and computer vision. He is currently part of the FakeID project in which he works on the development of AI-based methods for the generation and robust detection of Deepfakes. Likewise, he supports the NeuroHum project in acquiring a deeper understanding of humans perception of virtual avatars.
For further details about his activities please check the subsections below.
Biography
Education:
| 10/2018 - 03/2021 | M.Sc in Mathematics with minor in Computer Science at the University of Muenster |
| 03/2021 | Master Thesis: "A Robbins-Monro procedure for estimation in semiparametric models" |
| 01/2020 - 05/2020 | Semester abroad at the University of Gothenburg and Chalmers Institute of Technology in Gothenburg, Sweden |
| 10/2014 - 09/2017 | B.Sc in Mathematics with minor in Philosophy at the University of Muenster |
Work and Research Experience:
| since 08/2021 | Research associate, Computer Vision & Graphics Group, Fraunhofer HHI |
| 10/2020 - 03/2021 | Student assistant, Institute for Mathematical Stochastics, University of Muenster |
| 06/2020 - 10/2020 | Working student in learning experience at edyoucated GmbH |
| 06/2018 - 01/2020 | Working student in data science at Westphalia DataLab GmbH |
| 02/2018 - 05/2018 | Data Science Intern at Daimler AG |
Projects
FakeID
April 2021 - March 2024
Identification of Deepfakes in video-based authentication processes.
NeuroHum
May 2021 - April 2025
Neurocognitively-guided modelling of virtual humans for enhanced realism in immersive media.