Three Papers Accepted at DAGM-GCPR 2023

Three papers of our group have been accepted for publication at the DAGM German Conference for Pattern Recognition 2023:

• Dimitri Korsch, Maha Shadaydeh and Joachim Denzler:
  Simplified Concrete Dropout – Improving the Generation of Attribution Masks for Fine-grained Classification
  
• Jan Blunk, Niklas Penzel, Paul Bodesheim and Joachim Denzler:
  Beyond Debiasing: Actively Steering Feature Selection via Loss Regularization
  
• Matthias Körschens, Solveig Franziska Bucher, Christine Römermann and Joachim Denzler:
  Improving Data Efficiency for Plant Cover Prediction with Label Interpolation and Monte-Carlo Cropping

These works will be presented at the conference, which will be held September 19^th to September 22^nd, 2023 at Heidelberg, Germany. You can find pre-prints of these papers in our publication section.