We focus on machine-learning based approaches to computer vision. We are particularly interested in approaches that learn from data and that model spatio-temporal relations. We use our approaches to solve key computer vision problems such as:

• Representation Learning
• Cell biology - we are developing computer vision and machine learning methods for automated cell applications high-throughput live-cell imaging of Cell's dynamic processes
• Object detection and tracking. We apply our approaches in a variety of domains, including surveillance and human-robot interaction