Tim Büchner

Computer Vision Group Jena, Research Associate, PhD Student

prof_pic.jpg

I am currently doing my PhD at the Computer Vision Group Jena at the Friedrich Schiller University Jena under Prof. Dr.-Ing. Joachim Denzler. My research focuses on developing new algorithms, especially for facial expression analysis in medical and psychological applications in the context of muscular activity. My focus is not only on 2D images but also 3D data, going into 4D data.

In the DFG project Bridging the Gap: Mimics and Muscles, we aim to combine 3D surface changes with the underlying muscle activity to understand facial expressions better. The project is a collaboration between the Computer Vision Group Jena and the Ear-Nose-Throat Clinic Jena under Prof. Dr. Orlando Guntinas-Lichius. Together, we combine our computer vision and medical science expertise to further the understanding of facial expressions for patients with facial palsy.

Another area I am currently investigating is understanding the black-box nature of deep learning models, especially in medical applications. We investigate whether human-defined properties (such as facial symmetry) significantly influence the internally learned decision rules of the model. Our goal is to make non-experts understand the decision-making process of deep learning models and improve their trust in the model’s predictions, especially in out-of-distribution scenarios.

selected publications

  1. buechner2023improved
    Improved Obstructed Facial Feature Reconstruction for Emotion Recognition with Minimal Change CycleGANs
    In Advanced Concepts for Intelligent Vision Systems (Acivs) , 2023
    Best Paper Award
  2. buechner2023lets
    Let’s Get the FACS Straight - Reconstructing Obstructed Facial Features
    In International Conference on Computer Vision Theory and Applications (VISAPP) , 2023
  3. buechner2024power
    The Power of Properties: Uncovering the Influential Factors in Emotion Classification
    In International Conference on Pattern Recognition and Artificial Intelligence (ICPRAI) , 2024
    (accepted)
  4. vemuri2024functional
    Functional Tensor Decompositions for Physics-Informed Neural Networks
    Sai Karthikeya VemuriTim Büchner, Julia Niebling, and Joachim Denzler
    In International Conference on Pattern Recognition (ICPR) , Dec 2024
    (accepted)