Career Profile

Hi! My name is Florian Blume and I am a fourth-year PhD student at Technical University of Berlin in Olaf Hellwich’s computer vision lab. My research is focused on better alining human perception and synthetic processing in the case of facial expression recognition. We perceive in context and draw on prior knowledge that we have acquired but artificial systems largely do not. In my PhD, I lay the foundation for a research path that focuses on bridging this gap. I am currently writing up my thesis, if you are interested in my research or have job opportunities, hit me up!

Publications

I was first (or shared-first) author of the following publications:

  • How Do You Perceive My Face? Recognizing Facial Expressions in Multi-Modal Context by Modeling Mental Representations
  • Florian Blume, Runfeng Qu, Pia Bideau, Martin Maier, Rasha Abdel Rahman, Olaf Hellwich
    Accepted to GCPR 2024
  • Multi-Task Multi-Modal Self-Supervised Learning for Facial Expression Recognition
  • Marah Halawa, Florian Blume, Pia Bideau, Martin Maier, Rasha Abdel Rahman, Olaf Hellwich
    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024, pp. 4604-4614
  • Knowledge-augmented face perception: Prospects for the Bayesian brain-framework to align AI and human vision
  • Martin Maier, Florian Blume, Pia Bideau, Olaf Hellwich, Rasha Abdel Rahman
    Consciousness and Cognition, Volume 101, 2022, ISSN 1053-8100

    Education

    PhD in computer Science

    2020 (ongoing)
    Technical University of Berlin

    Focus on

    • Facial expression recognition in context
    • Computer vision
    • Interdisciplinary work with psychologists

    Diploma in Computer Science

    2012 - 2019
    Technical University of Dresden

    Grade: very good (1.4) Focus on

    • Machine learning
    • Computer vision
    • Theoretical computer science

    BSc in Computer Science

    2011 - 2012
    Karlsruhe Institute of Technology

    Projects

    Some of my open source projects on GitHub:

    6D-PAT - With 6D-PAT you can create 6D annotations on images for 6D pose estimation, i.e. annotate 2D images with the 3D rotation and 3D translation of 3D models.
    ConCluGen - Code for our paper "Multi-Task Multi-Modal Self-Supervised Learning for Facial Expression Recognition"
    Qt3D Offscreen Renderer - Implementation of an offscreen renderer for the Qt3D framework.
    Qt3D-Widget - Implementation of a Qt widget that allows to embed Qt3D.
    Qt3D Gizmo - A 3D gizmo implemented for the Qt3D framework.

    Work Experience

    Research Assistant

    05/2019 - 09/2019
    Max Planck institute für molecular cell biology, Dresden

    Development of an application to create mobile machine control apps.

    Student Employee

    02/2017 - 12/2017
    Monkey Works GmbH, Dresden

    Development of an application to create mobile machine control apps.

    Research Assistant

    03/2017 - 09/2017
    Computer lab of the faculty of law, Dresden

    Assisting in running the computer lab.

    Student Employee

    09/2015 - 12/2016
    exedio GmbH, Dresden

    Java-based backend development.

    Student Employee

    03/2014 - 08/2015
    Konsole Labs GmbH, Dresden

    Development of iOS radio apps.

    Awards

    Deutschlandstipendium

    2015 - 2016

    Receiver of the Deutschlandstipendium scholarship for talented students for one year.

    Skills & Proficiency

    Python & Pytorch

    Tensorflow

    C++

    Illustrator & Inkscape

    Java

    HTML5 & CSS