LAB VISION
The Computer Vision Lab (CVLab) at the University of Twente advances how intelligent systems perceive, model, and reason about the visual world. We bring together expertise in robust visual perception, multimodal learning, efficient learning, model interpretability, and video understanding.
Our vision is to develop visual AI that understands complex environments and human activity, adapts to changing conditions, and supports people in meaningful ways. We pursue foundational research in computer vision and machine learning, connecting scientific advances with the demands of real-world applications.
MISSION & MOTIVATION
Our mission is to develop visual AI systems that are capable, reliable, and understandable. We investigate how models can generalize beyond their training conditions, combine complementary information across modalities, learn efficiently, and capture how actions and interactions unfold over time. We also develop methods to examine model behavior, identify sources of bias, and make predictions and limitations easier to understand.
These challenges are closely connected. Understanding human behavior requires both visual and temporal context; learning from multiple modalities raises questions about robustness and efficiency; and applying models in practice requires understanding when and why they succeed or fail. By bringing these perspectives together, CVLab addresses research questions that span traditional boundaries within computer vision and machine learning.
Within the Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS), we contribute to the University of Twente’s ambition to put people at the center of technology. We work with academic, industry, and clinical partners in the Netherlands and internationally to develop research informed by practical needs.
CVLab is co-led by Nicola Strisciuglio, Estefanía Talavera, and Alex Stergiou. Through open science, collaborative research, and the education and mentorship of emerging researchers, we work towards visual AI that serves people well.