Instructors: Prof. Dr. Nassir NavabDr. Shahrooz Faghihroohi, Azade Farshad, Yousef Yeganeh

Time: TBA



  • The presentation and blogpost guidelines are available here: Guide_DLMA SS2022.pdf 
  • The preliminary meeting slides can be found here:
  • The preliminary meeting is scheduled for Feb 2nd, 10:30 (Zoom link is visible on TUMonline).


  • Deep Learning is growing tremendously in Computer Vision and Medical Imaging as well. Highly impacted journals in the medical imaging community, i.e. IEEE Transaction on Medical Imaging, recently published their special edition on Deep Learning [1]. The Seminar will propose a list of recent scientific articles related to the main current research topics in deep learning for Medical Applications, together with some interesting papers from other communities (CVPR, NeurIPS, ICCV, ICLR, ICML, ...).

Course Structure

In this Master Seminar (Hauptseminar), students select one scientific topic from the list provided by course organizers. The students should read the proposed sample papers by the tutors, find the topic-related articles, summarize and compare them in their presentation and blogpost:

  • Presentation: The selected paper is presented to the other participants (Maximum 25 minutes presentation, 10 minutes questions). You can use the CAMP templates for PowerPoint TUM-Template.pptx.
  • Blog Post: A blog post of 2500-3000 words excluding references, should be submitted before the deadline. The blog post must include all references used and must be written completely in your own words. Copy and paste will not be tolerated.
  • Attendance: Participants have to participate actively in all seminar sessions. Each presentation is followed by a discussion, and everyone is encouraged to actively participate.

Submission Deadline: You have to submit the blog post by the end of July 23rd.

Schedule (TBA)

DateSession: TopicsSlidesStudents

Recent Trends in Medical Image Segmentation

3D vessel segmentation

Structural Continuity in Segmentation

Huang, Pei-Ran

Sauer, Bjarne

Güvercin, Göktug


Exploring Latest Unsupervised Computer Vision Models for Segmentation

Self-supervised Volume Segmentation

Self-supervised graph representation learning

Klausen, Tobias

Altunbas, Begüm

Oytun Demirbilek


Image Superresolution Using Generative Models

Sound and Music Generative Models

Sensorless US compounding

Schauer, Robert

Victor Dzhagatspanyan

Sharma, Devansh


Application of Diffusion Models for Medical Imaging

Image to image translation with diffusion models

Sampling Methods in Diffusion Models

Cheng, JiaJian

Trigui Amal

Yeşilkaynak, Vahit Buğra


Converting weights of 2D Vision Transformer for 3D Image Classification

Natural Language Explanations for Vision and Vision-Language tasks

non-rigid 2d-3d registration

Image Stitching Using Unsupervised/Semi-Supervised Learning

Ben Chaaben, Zeineb

Marin Ruiz, Jorge

Yang, Shucheng

Güven Erkaya


Physics-inspired Neural Networks

Counterfactual Modelling

Confidence segmentation

Wagner, Jakob

Pennig, Lars

Yakal, Furkan


List of Topics and Material

The proposed papers for each topic in this course are usually selected from the following venues/publications:

CVPR: Conference on Computer Vision and Pattern Recognition
ICLR: International Conference on Learning Representations
NeurIPS: Neural Information Processing Systems

TPAMI: IEEE Transactions on Pattern Analysis and Machine Intelligence

TMI: IEEE Transaction on Medical Imaging
JBHI: IEEE Journal of Biomedical and Health Informatics
MedIA: Medical Image Analysis (Elsevier)

MICCAI: Medical Image Computing and Computer-Assisted Intervention
BMVC: British Machine Vision Conference
MIDL: Medical Imaging with Deep Learning

List of papers (TBA)

NoTopicSample PapersJournal/ ConferenceTutorStudentLink
1Counterfactual ModellingCOUNTERFACTUAL GENERATIVE NETWORKSICLR 2021Pennig, Lars
Uncertainty Estimation and Out-of-Distribution Detection for Counterfactual Explanations: Pitfalls and SolutionsICML 2021
ACAT: Adversarial Counterfactual Attention for Classification and Detection in Medical ImagingArXiv 2023
2Sampling Methods in Diffusion ModelsFast Sampling of Diffusion Models with Exponential IntegratorICLR 2023Yeşilkaynak, Vahit Buğra
3Image Stitching Using Unsupervised/Semi-Supervised LearningDepth-Aware Multi-Grid Deep Homography Estimation with Contextual CorrelationIEEE Transactions on CSVT 2022Shahrooz Faghihroohi Güven Erkaya
Unsupervised Deep Image Stitching: Reconstructing Stitched Features to ImagesTIP 2021
Semi-supervised Deep Large-baseline Homography Estimation with Progressive Equivalence ConstraintAAAI 2023
4Image Superresolution Using Generative ModelsDeep Constrained Least Squares for Blind Image Super-ResolutionCVPR 2022Shahrooz Faghihroohi Schauer, Robert

Progressive Residual Learning with Memory Upgrade for Ultrasound Image Blind Super-resolutionIEEE BHI 2022
Blind Image Super-Resolution: A Survey and BeyondPAMI 2023
5Sound and Music Generative ModelsNoise2Music: Text-conditioned Music Generation with Diffusion ModelsArXive 2023Victor Dzhagatspanyan
AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking HeadArXive 2023
AudioGen: Textually Guided Audio GenerationICLR 2023
63D vessel segmentation3D vessel-like structure segmentation in medical images by an edge-reinforced networkMedIASauer, Bjarne
3D Graph-Connectivity Constrained Network for Hepatic Vessel SegmentationIEEE JBHI
Noisy Labels are Treasure: Mean-Teacher-Assisted Confident Learning for Hepatic Vessel SegmentationMICCAI 2021
7Sensorless US compounding3D freehand ultrasound without external tracking using deep learningMedIA 2018Mohammad Farid Azampour Sharma, Devansh
Development of Implicit Representation Method for Freehand 3D Ultrasound Image Reconstruction of Carotid VesselIUS 2022
RecON: Online learning for sensorless freehand 3D ultrasound reconstructionMedIA 2023
8Image to image translation with diffusion modelsDUAL DIFFUSION IMPLICIT BRIDGES FOR IMAGE-TO-IMAGE TRANSLATIONICLR 2023Trigui Amal
Palette: Image-to-Image Diffusion ModelsArXiv 2022
9Structural Continuity in SegmentationDirectional Connectivity-based Segmentation of Medical ImagesCVPR 2023Güvercin, Göktug
Introducing Soft Topology Constraints in Deep Learning-based Segmentation using Projected Pooling LossSPIE Medical Imaging 2023
Exploring Discontinuity for Video Frame InterpolationCVPR 2023
10Self-supervised Volume SegmentationMasked Supervised Learning for Semantic SegmentationBMVC 2022Altunbas, Begüm
Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI ImagesMICCAI 2021
Volumetric Optimal Transportation by Fast Fourier TransformICLR 2023
11Physics-inspired Neural NetworksPhysics-Driven Diffusion Models for Impact Sound Synthesis from VideosCVPR 2023Wagner, Jakob
Phase2vec: dynamical systems embedding with a physics-informed convolutional networkICLR 2023
DaxBench: Benchmarking Deformable Object Manipulation with Differentiable PhysicsICLR 2023
12Exploring Latest Unsupervised Computer Vision Models for SegmentationEmerging Properties in Self-Supervised Vision TransformersArXiv 2023Klausen, Tobias
DINOv2: Learning Robust Visual Features without SupervisionArXiv 2023
Segment AnythingArXiv 2023
13Recent Trends in Medical Image SegmentationAttention-enhanced Disentangled Representation Learning for Unsupervised Domain Adaptation in Cardiac SegmentationMICCAI 2022Unknown User (ge94wiy) Huang, Pei-Ran

CRISP- Reliable Uncertainty Estimation for Medical Image SegmentationMICCAI 2022

Domain Specific Convolution and High Frequency Reconstruction Based Unsupervised Domain Adaptation for Medical Image Segmentation

14non-rigid 2d-3d registration A Weakly Supervised Framework for 2D/3D Vascular Registration Oriented to Incomplete 2D Blood VesselsIEEE Transactions on Medical Robotics and Bionics. 2022Yang, Shucheng

Non-rigid registration based on hierarchical deformation of coronary arteries in CCTA imagesBiomedical Engineering Letters. 2023
CNN-based real-time 2D-3D deformable registration from a single X-ray projectionArXiv 2022
15Converting weights of 2D Vision Transformer for 3D Image ClassificationCan We Solve 3D Vision Tasks Starting from A 2D Vision Transformer?ArxivMatthias KeicherBen Chaaben, Zeineb

[2209.07026] Can We Solve 3D Vision Tasks Starting from A 2D Vision Transformer? (

Adapting Pre-trained Vision Transformers from 2D to 3D through Weight Inflation Improves Medical Image SegmentationArxiv

[2302.04303] Adapting Pre-trained Vision Transformers from 2D to 3D through Weight Inflation Improves Medical Image Segmentation (

COVID Detection and Severity Prediction with 3D-ConvNeXt and Custom PretrainingsECCV 22 Workshop

COVID Detection and Severity Prediction with 3D-ConvNeXt and Custom Pretrainings | SpringerLink

16Learning-based Statistical Shape ModelDeep implicit statistical shape models for 3d medical image delineationAAAI 2022None
Deep Structural Causal Shape ModelsECCV 2022
Leveraging unsupervised image registration for discovery of landmark shape descriptorMedIA 2021
17Application of Diffusion Models for Medical ImagingUnsupervised Denoising of Retinal OCT with Diffusion Probabilistic ModelSPIE Medical Imaging 2022Cheng, JiaJian
On Conditioning the Input Noise for Controlled Image Generation with Diffusion ModelsArxiv
Fast Unsupervised Brain Anomaly Detection and Segmentation with Diffusion ModelsMICCAI 2022
18Self-supervised graph representation learningPrototype-based Embedding Network for Scene Graph GenerationCVPR 2023Oytun Demirbilek
Unbiased Scene Graph Generation in VideosCVPR 2023
Multi-task Self-supervised Graph Neural Networks Enable Stronger Task GeneralizationICLR 2023
19Natural Language Explanations for Vision and Vision-Language tasksNLX-GPT: A Model for Natural Language Explanations in Vision and Vision-Language TasksCVPR 2022David Bani-HarouniMarin Ruiz, Jorge

CLEVR-X: A Visual Reasoning Dataset for Natural Language ExplanationsICML 2022 workshop
ALICE: Active Learning with Contrastive Natural Language ExplanationsEMNLP 2020
20Confidence segmentationAutomated and real-time segmentation of suspicious breast masses using convolutional neural networkPloS one, 2018Yakal, Furkan
Leveraging Uncertainty Estimates to Improve Segmentation Performance in Cardiac MRMiccai 2021

Literature and Helpful Links

A lot of scientific publications can be found online.

The following list may help you to find some further information on your particular topic:

Some publishers:

Libraries (online and offline):

Some further hints for working with references:

  • JabRef is a Java program for comfortable working with Bibtex literature databases. Handy feature: if you know the PubMed ID for an article, JabRef can import data from there (via "Web Search/Medline").
  • Mendeley is a cross-platform program for organising your references.

If you find useful resources that are not already listed here, please tell us, so we can add them for others. Thanks.

  • No labels