• Tentative Topics
    • Self-supervised domain adaptation for historical documents using deep learning (CNNs) (Master)

     

    ... and more! Check out our research topics and research projects!

  • Ongoing Master Projects
    • Salim Hermidas - Reimplementation and Extension of DivaServices
    • Alain Schaller - Sterile-Zone Perturbations Recognition
    • Mattia Pedrazzi - Establishing a CNN baseline for a pT1 (and potentially stage II) CRC cohort dataset
    • Atefeh Fakhari - Applying The TEA-Graph Based Analysis to Colorectal Cancer Data
  • Ongoing Bachelor Projects
  • Finished Projects (Selection)

    Bachelor:

    • Gwenael Gendre - Graph-Based Keyword Spotting in Ancient Greek Papyri
    • Fabrice Prince - Deep Learning Approach for Epithelium Segmentation in Cancer Image Analysis
    • Pablo Wey - Graph Matching for Ground Truthing Chu Nom Characters
    • Peguy Njoyim Tchoubith - Machine learning experiments in document image analysis
    • Yannis Laaroussi - Data processing and graph visualisation of tumour and T-cell detections
    • Max Haller - Ground truth refinement for DIVAHisDB

    Master:

    • Carl Balmer - Tackling the Hanabi Challenge: Reinforcement Learning for a Cooperative Card Game
    • Joel Niklaus - Building a Super-Human Artificial Intelligence for the Swiss Card Game Jass
    • Dominic Kohler - Popularity Prediction of Reddit Posts
    • Jannis Wallau - Classification of Intestinal Glands using Graph Neural Networks
    • Sven Wallau - Transfer Learning for Intestinal Gland Segmentation
    • Manuel Drazyk - OCR Constrained Historical Document Synthesis with GANs
    • Matteo Badarocco - A POS Tagger Comparison between Prose Text and Poetry
    • Marion Schöneweid - Graph representations for H&E Whole Slide Images of Colorectal Cancer
    • Jonas Diesbach - Synthetization of Nom characters
    • Tanja Küry - Preserving Biodiversity in the Chinko Natural Reserve
    • Gwenael Gendre - Character Spotting in Ancient Greek Papyrus