To the OTH Regensburg website

Download Directory

  • Regensburg Breast Shape Model (RBSM)

    Statistical shape modeling of female breasts
    - Creation of statistical models of the female breast and development of new methods for 3D reconstruction of the breast from 2D images


  • SemiSup

    SMART-GI: AI projects for computer support in endoscopy of the upper gastrointestinal tract
    - Deep learning for early detection of esophageal cancer from endoscopic color images


  • The focus is on out-of-distribution (OOD) detection as a central mechanism for ensuring the reliability of AI systems in medical image processing. The aim is to develop and evaluate methods that reliably identify unknown or irregular inputs and thus contribute to patient safety.

    OpenMIBOOD Github repository: https://github.com/remic-othr/OpenMIBOOD/tree/main

    CVPR Paper Link: https: //cvpr.thecvf.com/virtual/2025/poster/33159


  • The reference implementation of the DIY Challenge Blueprint - a complete guide for building, hosting, and running biomedical image analysis challenges.

    DIY Challenge Blueprint Github repository: https: //github.com/remic-othr/PhaKIR_DIY

    MICCAI Paper Link: https: //papers.miccai.org/miccai-2025/paper/2800_paper.pdf