Index

Developing and Evaluating Image Similarity Metrics for Enhanced Classification Performance in 2D Datasets

Work description
This thesis focuses on the development and evaluation of novel image similarity metrics tailored for 2D datasets, aiming to improve the effectiveness of classification algorithms. By integrating active learning methods, the research seeks to refine these metrics dynamically through iterative feedback and validation. The work involves extensive testing and validation across diverse 2D image datasets, ensuring robustness and applicability in varied scenarios.

The following questions should be considered:

  • What metrics can effectively quantify the variance in a training dataset?
  • How does the variance within a training set impact the neural network’s ability to generalize to new, unseen data?
  • What is the optimal balance of diversity and specificity in a training dataset to maximize NN performance?
  • How can training datasets be curated to include a beneficial level of variance without compromising the quality of the neural network’s output?
  • What methodologies can be implemented to systematically adjust the variance in training data and evaluate its impact on NN generalization?

Prerequisites
Applicants should have a solid background in machine learning and deep learning, with strong technical skills in Python and experience with PyTorch. Candidates should also possess the capability to work independently and have a keen interest in exploring the theoretical aspects of neural network training.

For your application, please send your transcript of record.

Detectability Index Reimplementation for CT Images Using PyTorch

Work description
This project focuses on reimplementing the Detectability Index for evaluating individual CT projections, with the goal of improving the performance and adaptability of existing Python-based algorithms using PyTorch. The selected candidate will delve into the current code, identify performance bottlenecks, and propose innovative solutions to optimize efficiency. The goal is to minimize package dependencies to ensure code longevity and maintainability.

The following questions should be considered:

  • How can the existing Python code be improved with PyTorch for better performance and adaptability?
  • Where do the current code’s performance bottlenecks lie, and how can these be addressed?
  • How can the usage of external packages be minimized to ensure the code’s longevity?
  • What innovative approaches can be implemented to enhance the Detectability Index calculation?
  • How can the updated algorithm be validated for effectiveness and efficiency?

 

Prerequisites
Candidates should possess strong skills in Python and PyTorch, with the ability to quickly understand and improve upon existing code. A background in computational imaging or related fields, along with a problem-solving mindset, is essential.

For your application, please send your transcript of record.

Deep Learning-Driven Approaches for Optimizing Accuracy and Inference Speed in Compact Segmentation Models on Edge Devices

Automatic detection of Bronchoscopes on x-ray images

Scamming Scammers using Large Language Models

This Master Thesis is a cooperation with the Chair of Applied Cryptography.

Work description
In the digital age, scam emails have become a serious threat. These fraudulent emails aim to steal sensitive information or cause financial damage. This thesis aims to better understand the problem of scam emails and develop effective solutions to reduce their success. We will address several aspects, including the vulnerability of email addresses to scammers, the differentiation of scam emails from other dubious messages, the automation of responses through Large Language Models (LLMs), the detection of the usage of LLMs by the scammers, and the evaluation of the economic damage to the scammers based on the data obtained. We aim to strengthen the security of digital communication and help minimize the risks for users and organizations.

The following questions should be considered:

  • How can an email address be made vulnerable to scammers?
  • How can emails from scammers be distinguished from other dubious emails?
  • How can LLM responses be automated and customized?
  • How quickly do scammers recognize automated responses?
  • How can we accurately assess the extent of the economic harm caused by the scammer using our collected data?

 

Prerequisites
Prerequisites for this task include good knowledge of Deep Learning and IT Security, familiarity with Python and PyTorch, and the capability to work independently.

For your application, please send your transcript of record.

Quantification of Metal Artifacts in Metal Artifact Avoidance

AI-based Pavement Recognition System for Vehicle Road Infrastructure

AI-Driven Monuments Identification System and its details

In the realm of computer vision, significant research is currently dedicated to object detection
and recognition. Research groups and developers are actively striving to enhance machine
learning solutions, aiming to boost the accuracy of image detection and recognition in
accordance with specific use cases. The German AI-driven Monuments Detection System is an
innovative project aimed at providing tourists with an enhanced experience by leveraging
artificial intelligence (AI) to recognize and provide historical information about prominent
monuments in Germany. This document presents the project’s methodology, results, and
implications. With this machine, a user can scan the historical place and view its historical
details. There are five categories used in this model. Python is used as a programming
language with the TensorFlow framework.

Synthetic data creation of defect images for CNN training using GAN

Master Thesis in Cooperation with Infineon Technologies AG

External Supervision:
Weichselbaumer Christoph (BE R UPE TEST) (Christoph.Weichselbaumer@infineon.com)

Working Tittle:
Synthetic data creation of defect images for CNN training using GAN

Research Question:
Can the use of Generative Adversarial Networks (GANs) for generating synthetic data of particles and scratches on a transparent background improve the accuracy of Convolutional Neural Networks (CNNs) in defect detection?

Description:
The proposed Master’s thesis aims to address the challenge of providing ground truth for model training in AI use-cases, specifically in the context of detecting defects in products. Currently, the process of manual review and labeling of images for CNN [1] training is highly time-consuming and costly, and the most critical and relevant defects are often the least present in real data due to the design of products to minimize such defects. Moreover, data shifts in production can also affect the training of models.
To overcome these challenges, the proposed Master’s thesis will focus on the use of Generative Adversarial Networks (GANs) [2],[3], [4] to generate synthetic data for the minority classes for CNN training. The goal of the thesis is to create one or more GANs that are capable of generating defect images of particles and scratches on a transparent background, and to evaluate the performance of these generated images by measuring their classification accuracy using an existing CNN [5].

The proposed research will involve two sub-targets:
1. Developing GANs capable of generating defect images of particles and scratches on a transparent background.
2. Creating GANs capable of image-to-image translation to generate defect images of particles and scratches.
The focus of the research will be on generating defect images on a transparent background, with image-to-image translation considered an optional target depending on the progress of the research.

The approach to using GANs involves using existing defect image datasets for training, and then generating synthetic images to fill gaps in the real data. The GANs will be trained to generate synthetic images of particles and scratches on a transparent background, and these images will be combined with a “golden die” background to create an enhanced dataset for CNN training.
The suggested topic has the potential to significantly reduce the time and cost associated with manual review and labeling of images for CNN training, and to provide a more diverse and relevant dataset for model training. The proposed approach can also be applied to other defect types, such as stains and chipping, to further enhance the dataset for CNN training.

Literature:
[1]. O’Shea, Keiron, and Ryan Nash. “An introduction to convolutional neural networks.” arXiv preprint arXiv:1511.08458 (2015).
[2]. Ali, Safinah, Daniella DiPaola, and Cynthia Breazeal. “What are GANs? introducing generative adversarial networks to middle school students.” Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35. No. 17. 2021.
[3]. Figueira, Alvaro, and Bruno Vaz. “Survey on synthetic data generation, evaluation methods and GANs.” Mathematics 10.15 (2022): 2733.
[4]. Eilertsen, Gabriel, et al. “Ensembles of GANs for synthetic training data generation.” arXiv preprint arXiv:2104.11797 (2021).
[5]. Buda, Mateusz, Atsuto Maki, and Maciej A. Mazurowski. “A systematic study of the class imbalance problem in convolutional neural networks.” Neural networks 106 (2018): 249-259.

End-to-end detection and 3D localization of implants from multi-view images for surgical CBCT metal artifact avoidance

This project investigates the feasibility of directly regressing pose and position of metallic implants from multiple calibrated X-Ray images. In the context of intraoperative Cone-beam Computed Tomography (CBCT) imaging, this information can be used to avoid metal artifacts by adapting the scanning trajectory such as discussed in this previous work [1].

By formulating the problem as a set prediction problem, we can build on previous works such as DETR [1] to design an algorithm which directly models the depicted metallic objects. In addition to applying these existing works, which were developed for and tested on day-to-day optical images, we assess the possibility of incorporating additional knowledge about the relative geometry between the X-Ray images into the model architecture.

 

[1] Cone-beam CT trajectory optimization for metal artifact avoidance using ellipsoidal object parameterizations (spiedigitallibrary.org)
[2] [2005.12872] End-to-End Object Detection with Transformers (arxiv.org)