Computer Vision and Biometrics

• Computer Vision and Biometrics
• [B] Morphing robust face recognition
• [B] [M] FISWG AUTO: Automatic extraction and use of forensic facial features
• [B] Analyzing Visual and Identity-Level Differences Between Synthetic and Real Face Datasets
• [B] Heritage biometrics – 3D acquisition design for fingerprints, scratches and toolmarks on works of art
• [MB] Will it fail? Generating videos of actions fails
• [MB] Deriving text descriptions from visual model's predictions
• [B] Heritage Biometrics – Possibilities with AI in Furniture & Wood Conservation
• [B] Heritage Biometrics: does every painter paint himself?
• [B] Heritage biometrics – artist profiling from 3D fingerprints
• [B] Temporal Surgical Video Segmentation for Medical Dataset Curation
• [B] Surgical Phase Pattern Recognition from Video Data
• [B] Driver Behavior Understanding for Monitoring Fatigue and Anomalous Behavior in Autonomous Driving
• [B] Discovery of missing objects in predefined scene images
• [B] Pose Estimation in Handball Videos for Personalized Training
• [B] Egocentric vision for human behaviour understanding
• [B] Understanding the Effect of Image Quality and Resolution on Skin Lesion Recognition
• [B] Recognition of Crops Relevant for Food Security from Images – Dataset Compilation and Classification Framework Design
• [B] Benchmarking Medical Class-OOD (aka Open Set Recognition) Detection
• [B] OOD detection on MediCLIP
• [B] Can Normalizing Disentangled Flows Detect OOD data?
• [B] Medical OOD detection with an Exponentially Tilted Gaussian Prior for beta-Variational Autoencoders
• [B] Can you find the Gorilla in the CT scan? (2 topics: Sample and Object level)
• [B] Why can’t Vision-Language Models like CLIP find OOD data in medical images? Or can they? How?
• [B] Can Normalizing Disentangled Flows Detect OOD data?
• [B] Disentangled Flows for Face editing
• [B] Disentangled Flows for Anomaly Detection
• [B] Self-Supervised Out-of-Distribution Detection
• [B] Unsupervised Anomaly Detection Under Domain Shifts
• [B] [M] AUTO: Auto encoders in face recognition
• [M] [B] Computer Vision & Machine Learning @ Lely
• [B] [M] DIET: from obese data to sparse face models
• [B] Evaluation of AI-driven interpretation on microscopic images in oncology
• [M][B] Visual place recognition
• [B] TURBULENCE: 2D gen – 3D pose pipeline – atmospheric turbulence

Data Engineering

• Data Engineering
• Analysing coaching and therapy reports with Natural Language Processing/LLMs (Dutch, CS&BIT)
• Text mining to retrieve suspected cause of acute kidney injury from medical files
• Lesion detection model to monitor MS progression in brain MRI scans
• Natural Language Processing for diagnosis of patients with unexplainable Left Ventricle Hypertrofia
• Process Mining Opportunities at Univé
• Lifestyle Data Futures

Machine Learning

• Machine Learning
• [M] Virtual symphony: matching off-tempo audio from different instruments
• [M] Uni-modal to multi-modal model merging
• [M] BBI Process Mining: Predict the remaining processing time of a pallet for an automated refrigerated warehouse
• [M] Machine learning to quantify e3value business models
• [M] Research into the AI-Revolution in Camera Technology at LOOGISCH
• [M] Process Mining Pancreascarcinoom
• [M] Event-streaming Algorithm for Self-adaptive Distributed Vehicle Route Guidance in Urban Networks
• [M] Comparing attack graphs with business process models for modeling attacker strategies
• [M] Optimisation of Complication Registration with Text Mining and Machine Learning
• [M] LegalTech: AI and Legal processes
• [M] Advancing Healthcare Process Simulation: Automating Discrete Event Models from Process Data
• [M] Algorithm Development for Analysis of Heart Tissue
• [M] Fitness prediction for hip surgery for elderly patients
• [M] Probability prediction for acute kidney injury
• [M] (Multimodal) Admission prediction for patients in the cardiac emergency department
• [M] Fast-track outcome prediction
• [M] Real-time medication monitoring for patients with chronic pulmonary conditions
• [M] Quality Control on radiology MR reports
• [M] NLP model to predict diabetes type and year of diagnosis from medical texts
• [M] Artificial Intelligence against Cancer
• [M] AERIAL IMAGE SEGMENTATION WITH DATA-EFFICIENT ADAPTATION OF FOUNDATION MODELS
• [M] Auto Data Evaluation using Machine Learning @ROSEN Enschede
• [M] Object Detection for Water Pipeline Inspection @ROSEN Enschede
• [M] Autoncoders and Recovering High Frequency Components
• [D][M] BIO marker-based COVID Severity prediction and data quality exploration
• [M] Decoding EMG using A Riemannian Geometry approach
• [M] Decoding EMG using Autoencoders
• [M] Explainable Machine Learning
• [M] Explaining Estrus Classifications
• [M] Large-scale data mining & NLP @ OCLC - Leiden, NL
• [M] Machine Learning
• [M] Machine Learning and Natural Language Processing Internship @Nedap
• [M] Deep learning in medical imaging project in Berlin using 3D-printed phantom
• [M][B] Digital media - using neural networks for steganography
• [M][B] Fourier-basis noise image data augmentation for CNN robustness
• [M][B] Deblurring and disentangled representations of cell images via deep learning
• [M] Detect and correct wrong transliterations
• [M][B] Deep learning techniques for deblurring cell images utilizing geometry
• [M] How do different down-sampling methods affect neural networks’ reaction towards different frequencies?
• [M] Motion magnification for engineered heart tissue microscopy videos
• [M] Learning temporally-extended state and action representations in Deep Reinforcement Learning
• [M] Automated Algorithm Configuration of Sparse Neural Networks
• [M] Mortality prediction after hip fracture surgery. Could Machine Learning add value?
• [M] Point Cloud Classification using Context based Feature Engineering
• [M] Project: E-Manager Chronic Diseases
• [M] Semi-Supervised Deep Learning for Point Cloud Classification
• [M] Serendipity at the MuseumFabriek
• [M] Using physiological signals to monitor mental state of Cybersecurity experts at Dutch Bank