Fabian Herzog

I am a research assistant and PhD student at the Institute of Human-Machine Communication of the Technical University of Munich, Germany. My main research is in computer vision and pattern recognition.

Prior to joining TU Munich, I did my Master's in Scientific Computing in Göttingen with a focus on mathematical image processing. I then worked for fino as a data scientist.

Email  /  Google Scholar  /  Github  /  Twitter  /  LinkedIn

profile photo

My research is on deep learning and pattern recognition for video processing in general and object tracking in particular. Here you can find my recent projects and papers. Contributions as first author are highlighted.

Dissected 3D CNNs: Temporal Skip Connections for Efficient Online Video Processing
Okan Köpüklü, Stefan Hörmann, Fabian Herzog, Hakan Cevikalp, and Gerhard Rigoll
arXiv, 2020
arXiv / code

We propose dissected 3D CNNs to address several serious handicaps of 3D ResNet models during online operation.

Comparative Analysis of CNN-based Spatio-Temporal Modeling
Okan Köpüklü, Fabian Herzog, and Gerhard Rigoll
arXiv, 2019
arXiv / code

A comparative analysis of different spatiotemporal modeling techniques for action and gesture recognition.


Winter 2020 / 2021

  • Practical Course on Audio and Image Processing

Summer 2020

  • Human-Machine Communication II (Exercises)
  • Musical Acoustics (Assistance) (canceled due to COVID-19)

Winter 2019 / 2020

  • Practical Course on Audio and Image Processing
  • Practical Course on Systems Engineering

Summer 2019

  • Human-Machine Communication II (Exercises)


I'm supervising theses in the areas of computer vision and pattern recognition. Due to the high number of applications, I can only respond to complete applications with current CV, transcript of records and motivation letter.
This website template was stolen with permission from Jon Barron. Imprint.