Christoph Reich

I'm an ELLIS Ph.D. student at the Technical University of Munich and TU Darmstadt with co-supervision from the University of Oxford. My research focuses on unsupervised scene understanding and self-supervised learning. I'm supervised by Daniel Cremers (CVG), Stefan Roth (VisInf), and Christian Rupprecht (VGG). Prior to my PhD, I worked as a research intern at NEC Laboratories America (Princeton). Together with my supervisor Biplob Debnath, I worked on controlling standard video codecs for current computer vision models using self-supervised learning.

During my studies at TU Darmstadt, I work at the Self-Organizing Systems Lab (Heinz Koeppl) with Tim Prangemeier on 2D and 3D segmentation methods for biomedical applications and generative adversarial networks for live-cell in silico experiments. I also collaborated with the Artificial Intelligent Systems in Medicine Lab (Christoph Hoog Antink) focusing on ECG classification using deep learning.

I received a bachelor's in Information Systems Technology (computer science U electrical engineering) and a master's in Autonomous Systems (computer science) from TU Darmstadt. During my bachelor, I work as a working student at Bosch and as a teaching assistant at the mathematics department and the E5 Lab.

Email  /  CV  /  GitHub  /  Google Scholar  /  LinkedIn  /  Twitter

profile photo

News

  • 02/2024: 🚀 Happy to start my Ph.D. at CVG (TUM) and VisInf (TU Darmstadt) with co-supervision from VGG (University of Oxford).
  • 01/2024: 🥳 I was recognized as an outstanding reviewer at WACV 2024.
  • 08/2023: đź“‘ Our paper ''Differentiable JPEG: The Devil is in the Details'' has been accepted to WACV 2024.
  • 08/2023: đź“‘ Our paper ''The TYC Dataset for Understanding Instance-Level Semantics and Motions of Cells in Microstructures'' has been accepted to ICCVW 2023.
  • 04/2023: đź“‘ Two papers ''An Instance Segmentation Dataset of Yeast Cells in Microstructures'' and ''On the Atrial Fibrillation Detection Performance of ECG-DualNet'' has been accepted to EMBC 2023.
  • 07/2022: 👨‍🔧 Excited to start my research internship at NEC Laboratories America.
  • 06/2022: đź“‘ Our journal article has been accepted to appear in Physiological Measurement.
  • 06/2022: 🥳 I got awarded a Lightning AI Travel Scholarship to visit the 2022 Lightning AI DevCon in NYC.
  • 06/2022: 🥳 I got awarded the Papers with Code (Meta AI) Contributor Award 2022.
  • 03/2022: 👨‍💻 My Swin Transformer V2 reimplementation (with the amazing help of Ross Wightman) has been merged into the PyTorch Image Models library.
  • 10/2021: đź“‘ Our paper ''OSS-Net: Memory Efficient High Resolution Semantic Segmentation of 3D Medical Data'' has been accepted to BMVC 2021.
  • 10/2021: đź“‘ Our journal article has been accepted to appear in BioSystems (CIBCB special issue).
  • 06/2021: 🥳 My B.Sc. thesis has been awarded the Thesis-Award of the Centre for Synthetic Biology at TU Darmstadt.
  • 06/2021: đź“‘ Our paper ''Multi-StyleGAN: Towards Image-Based Simulation of Time-Lapse Live-Cell Microscopy'' has been accepted to MICCAI 2021.

Research

I'm interested in computer vision and deep learning. First author publications highlighted.

Deep Image Codec Control for Vision Models


Christoph Reich
Technische Universität Darmstadt, Visual Inference Lab, Master Thesis, 2023
arXiv    talk    slides    BibTeX   

-

Transformer Network with Time Prior for Predicting Clinical Outcome from EEG of Cardiac Arrest Patients


Maurice Rohr, Tobias Schilke, Laurent Willems, Christoph Reich, Sebastian Dill, Gökhan Güney, Christoph Hoog Antink
50th Computing in Cardiology Conference (CinC), 2023
paper    BibTeX   

-

Deep Video Codec Control for Vision Models


Christoph Reich, Biplob Debnath, Deep Patel, Tim Prangemeier, Daniel Cremers, Srimat Chakradhar
arXiv:2308.16215, 2023
arXiv    BibTeX   

-

Differentiable JPEG: The Devil is in the Details


Christoph Reich, Biplob Debnath, Deep Patel, Srimat Chakradhar
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024
paper    arXiv    project page    code    video    slides    poster    BibTeX   

-

The TYC Dataset for Understanding Instance-Level Semantics and Motions of Cells in Microstructures


Christoph Reich, Tim Prangemeier, Heinz Koeppl
IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2023
paper    arXiv    project page    code    slides    poster    dataset    BibTeX   

-

An Instance Segmentation Dataset of Yeast Cells in Microstructures


Christoph Reich*, Tim Prangemeier*, André O. Françani, Heinz Koeppl
45th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), full paper, 2023 (* equal contribution)
paper    arXiv    project page    code    slides    dataset    BibTeX   

-

On the Atrial Fibrillation Detection Performance of ECG-DualNet


Christoph Reich, Maurice Rohr, Tim Kircher, Christoph Hoog Antink
45th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Short Paper, 2023
paper    code    poster    BibTeX   

-

Exploring Novel Algorithms for Atrial Fibrillation Detection by Driving Graduate Level Education in Medical Machine Learning


Maurice Rohr, Christoph Reich, Andreas Höhl, Timm Lilienthal, Tizian Dege, Filip Plesinger, Veronika Bulkova, Gari D Clifford, Matthew A Reyna, Christoph Hoog Antink
Physiological Measurement, 2022
paper    code    BibTeX   

-

Histopathological Image Classification based on Self-Supervised Vision Transformer and Weak Labels


Ahmet Gokberk Gul*, Oezdemir Cetin*, Christoph Reich, Nadine Flinner, Tim Prangemeier, Heinz Koeppl
Medical Imaging: Digital and Computational Pathology, 2022 (* equal contribution)
paper    arXiv    code    BibTeX   

-

Yeast Cell Segmentation in Microstructured Environments with Deep Learning


Tim Prangemeier, Christian Wildner, André O. Françani, Christoph Reich, Heinz Koeppl
Biosystems (CIBCB special issue), 2022
paper    code    BibTeX   

-

OSS-Net: Memory Efficient High Resolution Semantic Segmentation of 3D Medical Data


Christoph Reich, Tim Prangemeier, Ă–zdemir Cetin, Heinz Koeppl
British Machine Vision Conference (BMVC), 2021
paper    arXiv    project page    code    video    slides    BibTeX   

-

Multi-StyleGAN: Towards Image-Based Simulation of Time-Lapse Live-Cell Microscopy


Christoph Reich*, Tim Prangemeier*, Christian Wildner, Heinz Koeppl
International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2021 (* equal contribution)
paper    supplement    arXiv    project page    code    slides    poster    dataset    BibTeX   

-

Generation and Simulation of Yeast Microscopy Imagery with Deep Learning


Christoph Reich
Technische Universität Darmstadt, Self-Organizing Systems Lab, Bachelor Thesis, 2020
arXiv    award    BibTeX   

2020 Thesis Award of the Centre for Synthetic Biology

Attention-Based Transformers for Instance Segmentation of Cells in Microstructures


Tim Prangemeier, Christoph Reich, Heinz Koeppl
IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2020
paper    arXiv    code    BibTeX   

-

Multiclass Yeast Segmentation in Microstructured Environments with Deep Learning


Tim Prangemeier, Christian Wildner, André O. Françani, Christoph Reich, Heinz Koeppl
IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2020
paper    arXiv    BibTeX   

-



Other project

These include side projects, open-source contributions, and coursework. Smaller projects can be found at my GitHub profile.

Talk - Attention, GPT, ViT, and ChatGPT


Technische Universität Darmstadt, Self-Organizing Systems Lab, 2023
slides    video   

Lab talk on Attention, GPT, ViT, and ChatGPT.

MaxViT: Multi-Axis Vision Transformer REIMPLEMENTATION


Private project, 2022
code   

Reimplementation of the paper “MaxViT: Multi-Axis Vision Transformer” by Zhengzhong Tu et al. (image taken from paper).

Lightweight Probabilistic Deep Networks for Cell Segmentation


Technische Universität Darmstadt, AI & ML Lab, 2022
slides    video   

Probabilistic graphical models course project supervised by Prof. Kristian Kersting. Utilizing Lightweight Probabilistic Deep Networks by Gast and Roth for trapped yeast cell segmentation.

Swin Transformer V2: Scaling Up Capacity and Resolution REIMPLEMENTATION


Private project, 2022
code   

Reimplementation of the paper “Swin Transformer V2: Scaling Up Capacity and Resolution” by Ze Liu et al. (image taken from paper). Code merged into the PyTorch image models library by Ross Wightman (release).

ECG-DualNet: Atrial Fibrillation Classification in Electrocardiography using Deep Learning


Technische Universität Darmstadt, KIS*MED, 2021
paper    code    slides   

Project seminar supervised by Prof. Christoph Hoog Antink and Maurice Rohr on classifying atrial fibrillation in ECG signals using Deep Learning

Dirac-GAN REIMPLEMENTATION


Technische Universität Darmstadt, MEC-Lab, 2021
code   

Course project supervised by Anirban Mukhopadhyay reimplementation of the Dirac-GAN proposed by Lars Mescheder et al. in the paper “Which Training Methods for GANs do actually Converge?” published at ICML 2018

Involution: Inverting the Inherence of Convolution for Visual Recognition REIMPLEMENTATION


Private project, 2021
code   

Reimplementation of the CVPR 2021 paper “Inverting the Inherence of Convolution for Visual Recognition” (2D and 3D Involution) by Duo Li et al. (image taken from paper)

Mode Collapse Example for Various GAN Losses


Private project, 2021
code   

Implementation of various GAN losses to showcase mode collapse in 2D

DeepFovea++: Reconstruction and Super-Resolution for Natural Foveated Rendered Videos


Technische Universität Darmstadt, AI & ML Lab, 2020
code    slides    report   

Deep learning course project supervised by Prof. Kristian Kersting (group project with Marius Memmel and Jonas Grebe)

ToeffiPy a PyTorch like autograd/deep learning library based on NumPy


Private project, 2020
code   

Building a PyTorch like autograd/deep learning library based only on NumPy

Semantic Pyramid for Image Generation PyTorch REIMPLEMENTATION


Private project, 2020
code   

Reimplementation of the CVPR 2020 paper “Semantic Pyramid for Image Generation” by Assaf Shocher et al.

Scalable 3D Semantic Segmentation for Gun Detection in CT Scans


Technische Universität Darmstadt, Visual Inference, 2020
arxiv    code    report   

Project lab deep learning in computer vision supervised by Prof. Stefan Roth and Faraz Saeedan (group project with Marius Memmel and Nicolas Wagner)

Convolutional Neural Network Application for Cell Segmentation in Biological Research


Technische Universität Darmstadt, Self-Organizing Systems Lab, 2019
report   

Project seminar supervised by Prof. Heinz Koeppl, Tim Prangemeier and Christian Wildner





Design / source code from Jon Barron's website


© Christoph Reich 2023