Teaching

Information for students who are interested in starting a BSc or MSc thesis.

• Thematically, you should be interested in image processing, compression, or the mathematical foundations of deep learning.
• You should have attended a suitable amount of topic-related lectures.
• You can propose your own topic ideas or ask for recommendations.
• Check study regulations and the PS Mint page for general information about BSc and MSc theses.
• Contact Pascal Peter for a initial meeting to verify that you fulfil all requirements and to discuss first topic propositions.

• Hevra Petekkaya: Mask Density Estimation with Deep Learning
• Beste Ekmen: Inpainting Artworks with Style in Mind
• Enes Ulus: Prediction Strategies for Inpainting-based Compression
• Jorge Augusto Calvimontes Robles: Bridging Deep Learning and Variational Models: Leveraged Loss and Parameter Optimization for Exposure Fusion
• Aseer Ahmad Ansari: Deep-Learning Mask Optimisation for Osmosis
• Moritz Altmeyer:  Coding Strategies for Inpainting-based Compression with Simple Ingredients

• Kevin Baum: GPGPU Supportet Diffusion-Based Naive Video Compression. M.Sc. Thesis in Computer Science, 2013.
• Alexander Scheer: Entropy Coding for PDE-Based Image Compression. B.Sc. Thesis in Computer Science, 2014.
• Zhao Jin: Video Coding with Three-Dimensional Diffusion. M.Sc. Thesis in Computers and Communications, 2014
• Frank Nedwed: A Probabilistic Approach to Image Compression Using Subdivisions. B.Sc. Thesis in Computer Science, 2015.
• Hui Men: Scene-Detection for Diffusion-Based Video Compression. M.Sc. Thesis in Computer Science, 2016.
• Marie Mühlhaus: Compressing Binary Images with the Medial Axis Transform. B.Sc. Thesis in Computer Science.
• Robin Dirk Adam: Denoising by Inpainting. M.Sc. Thesis in Computer Science, 2016.
• Sreenivas Narashima Murali: Non-Local Patch based Error Measure for Texture Classification. M.Sc. Thesis in Visual Computing, 2016.
• Jan Contelly: Audio Signal Compression with Inpainting Ideas. M.Sc. Thesis in Computer Science, 2016.
• Jillian Clark: Adaptive Quantisierung für PDE-basierte Bildkompression. Thesis in Computer Science for High School Teachers, 2018.
• Merlin Köhler: Learning Optimal Data for Sparse Image Representation. M.Sc. Thesis in Computer Science, 2018.
• Vincent Kübler: Inpainting-based Compression of Noisy Images. B.Sc. Thesis in Computer Science.
• Abhishekh Goswami: Inpainting-based Compression with Simple Ingredients. M.Sc. Thesis in Visual Computing, 2018.
• Lena Karos: Optimal Data Selection for Exemplar-Based Inpainting. M.Sc. Thesis in Computer Science, 2018.
• Niklas Kämper: Encoding Strategies for Pixel Data in PDE-based Compression. B.Sc. Thesis in Computer Science, 2019.
• Rahul Mohideen: Efficient Encoding Strategies for Inpainting-Based Compression with Exact Masks. M.Sc. Thesis in Visual Computing, 2019.
• Niklas Kämper: Neural Decoding for RJIP. M.Sc. Thesis in Computer Science, 2020.
• Moritz Kunz: Block-based Image Compression. M.Sc. Thesis in Mathematics, 2021.
• Yaroslav Mykoliv: Event Compression. M.Sc. Thesis in Computer Science, 2021.
• Huasheng Chen: Structure-guided Exemplare-based Inpainting. B.Sc. Thesis in Computer Science, 2022.
• Matthias Hock: End-to-end Image Compression with CNNs. M.Sc. Thesis in Computer Science, 2023.
• Paul Bungert: Imate Stitching with Osmosis. M.Sc. Thesis in Computer Science, 2023.
• Tom Fischer: Optical Flow with Explicit Diffusion. M.Sc. Thesis in Computer Science, 2023.
• Julia Gierke: Skeletonisation Scale-Spaces. B.Sc. Thesis in Mathematics, 2024.
• Huasheng Chen: Patch-based Inpainting with Neural Networks. M.Sc. Thesis in Computer Science, 2024.