Niklas Penzel, M.Sc.
Niklas Penzel
Address: Computer Vision Group
Department of Mathematics and Computer Science
Friedrich Schiller University of Jena
Ernst-Abbe-Platz 2
07743 Jena
Germany
Phone: +49 (0) 3641 9 46335
E-mail: niklas (dot) penzel (at) uni-jena (dot) de
Room: 1224
Links: Google Scholar
Curriculum Vitae
Since Dec. 2020 Research Associate at the Computer Vision Group, Friedrich Schiller University Jena
2020 Master Thesis: “The Bias Uncertainty Sampling introduces into an Active Learning System”
2018-2020 M.Sc. in Computer Science at the Friedrich Schiller University Jena
2018 Bachelor Thesis: “Lebenslanges Lernen von Klassifikationssystemen ohne Vorwissen und mit intelligenter Datenhaltung” (Lifelong Learning of Classification Systems without Previous Knowledge and with Smart Data Management)
2015-2018 B.Sc. in Computer Science at the Friedrich Schiller University Jena
Research Interests
  • Explainable AI
  • Analyzing Model Training
  • Lifelong Learning
  • Deep Learning
  • Super Resolution
Supervised Theses
  • Phillip Rothenbeck: “SIR-based modelling of COVID-19 pandemic using PINNs”. Bachelor thesis, 2024. (joint supervision with Sai Karthikeya Vemuri)
  • Maria Gogolev: “Comparing and Modifying Distributions of Latent Diffusion Models to Impose Image Properties”. Master thesis, 2024. (joint supervision with Sven Sickert and Tim Büchner)
  • Konstantin Roppel: “Model Feature Attribution for Single Images using Conditional Independence Tests”. Master thesis, 2024 (joint supervision with Jan Blunk)
  • Jan Blunk: “Steering Feature Usage During Neural Network Model Training”. Master thesis, 2023 (joint supervision with Paul Bodesheim)
  • Tristan Piater: “Self-Attention Mechanisms for the Classification of Dermoscopic Images”. Bachelor thesis, 2022 (joint supervision with Gideon Stein)
  • Maria Gogolev: “Continual fine-tuning with intelligent rehearsal selection”. Bachelor thesis, 2022 (joint supervision with Julia Böhlke)
Publications
2025
Niklas Penzel, Gideon Stein, Joachim Denzler:
Change Penalized Tuning to Reduce Pre-trained Biases.
Communications in Computer and Information Science. 2025. (in press)
[bibtex] [abstract]
Tristan Piater, Niklas Penzel, Gideon Stein, Joachim Denzler:
Self-Attention for Medical Imaging - On the need for evaluations beyond mere benchmarking.
Communications in Computer and Information Science. 2025. (in press)
[bibtex] [abstract]
2024
Niklas Penzel, Gideon Stein, Joachim Denzler:
Reducing Bias in Pre-trained Models by Tuning while Penalizing Change.
International Conference on Computer Vision Theory and Applications (VISAPP). Pages 90-101. 2024.
[bibtex] [web] [doi] [abstract]
Tim Büchner, Niklas Penzel, Orlando Guntinas-Lichius, Joachim Denzler:
Facing Asymmetry - Uncovering the Causal Link between Facial Symmetry and Expression Classifiers using Synthetic Interventions.
Asian Conference on Computer Vision (ACCV). 2024. (accepted at ACCV)
[bibtex] [pdf] [abstract]
Tim Büchner, Niklas Penzel, Orlando Guntinas-Lichius, Joachim Denzler:
The Power of Properties: Uncovering the Influential Factors in Emotion Classification.
International Conference on Pattern Recognition and Artificial Intelligence (ICPRAI). 2024.
[bibtex] [web] [doi] [abstract]
Tristan Piater, Niklas Penzel, Gideon Stein, Joachim Denzler:
When Medical Imaging Met Self-Attention: A Love Story That Didn’t Quite Work Out.
International Conference on Computer Vision Theory and Applications (VISAPP). Pages 149-158. 2024.
[bibtex] [web] [doi] [abstract]
2023
Jan Blunk, Niklas Penzel, Paul Bodesheim, Joachim Denzler:
Beyond Debiasing: Actively Steering Feature Selection via Loss Regularization.
DAGM German Conference on Pattern Recognition (DAGM-GCPR). 2023.
[bibtex] [pdf] [abstract]
Niklas Penzel, Jana Kierdorf, Ribana Roscher, Joachim Denzler:
Analyzing the Behavior of Cauliflower Harvest-Readiness Models by Investigating Feature Relevances.
ICCV Workshop on Computer Vision in Plant Phenotyping and Agriculture (CVPPA). Pages 572-581. 2023.
[bibtex] [pdf] [abstract]
Niklas Penzel, Joachim Denzler:
Interpreting Art by Leveraging Pre-Trained Models.
International Conference on Machine Vision and Applications (MVA). Pages 1-6. 2023.
[bibtex] [doi] [abstract]
2022
Niklas Penzel, Christian Reimers, Paul Bodesheim, Joachim Denzler:
Investigating Neural Network Training on a Feature Level using Conditional Independence.
ECCV Workshop on Causality in Vision (ECCV-WS). Pages 383-399. 2022.
[bibtex] [pdf] [doi] [abstract]
2021
Christian Reimers, Niklas Penzel, Paul Bodesheim, Jakob Runge, Joachim Denzler:
Conditional Dependence Tests Reveal the Usage of ABCD Rule Features and Bias Variables in Automatic Skin Lesion Classification.
CVPR ISIC Skin Image Analysis Workshop (CVPR-WS). Pages 1810-1819. 2021.
[bibtex] [pdf] [web] [abstract]
Niklas Penzel, Christian Reimers, Clemens-Alexander Brust, Joachim Denzler:
Investigating the Consistency of Uncertainty Sampling in Deep Active Learning.
DAGM German Conference on Pattern Recognition (DAGM-GCPR). Pages 159-173. 2021.
[bibtex] [pdf] [web] [doi] [abstract]