Philipp Harzig

Research Assistant & PhD Student
Chair for Machine Learning & Computer Vision
Phone: +49 (821) 598 4334
Room: 1024 (N)
Visiting hours: By appointment
Address: Universitätsstraße 6a, 86159 Augsburg

Research Interests

  • Computer Vision
  • Machine Learning
  • Visual Question Answering
  • Image Captioning
  • Deep Convolutional Neural Networks
  • Recurrent Neural Networks
  • Machine Learning Optimization


  • Automatic Image Captioning of Scenes Depicting Branded Products
  • Company Logo Detection



  • Philipp Harzig, Moritz Einfalt, Rainer Lienhart. 
    BioMedia Grand Challenge: Automatic Disease Detection and Report Generation for Gastrointestinal Tract Examinations. 
    ACM Multimedia (ACMMM) 2019. Nice, France, October 2019, to appear
  • Philipp Harzig, Yan-Ying Chen, Francine Chen, Rainer Lienhart. 
    Addressing Data Bias Problems for Chest X-ray Image Report Generation. 
    British Machine Vision Conference (BMVC) 2019. Cardiff, UK, September 2019,  [arXiv]
  • Philipp Harzig, Dan Zecha, Rainer Lienhart, Carolin Kaiser, René Schallner. 
    Image Captioning with Clause-Focused Metrics in a Multi-Modal Setting for Marketing.
    IEEE MIPR 2019 San José, CA, USA, March 2019 [ PDF]
  • Philipp Harzig, Christian Eggert, Rainer Lienhart. 
    Visual Question Answering With a Hybrid Convolution Recurrent Model.
    ACM International Conference on Multimedia Retrieval 2018 (ACM ICMR 2018) Yokohama, June 2018 [ PDF]
  • Philipp Harzig, Stephan Brehm, Rainer Lienhart, Carolin Kaiser, René Schallner. 
    Multimodal Image Captioning for Marketing Analysis.
    IEEE MIPR 2018 Miami, FL, USA, April 2018 [ PDF]

Master Thesis

Philipp Harzig.

Implementation of Frequency Domain Convolution for the Caffe-Framework.
Master Thesis, February 2016.