Dr. Tanja Teuber
Carl Zeiss AG

Title of the inspiring lecture | 11:40 PM – 12:00 PM
Bridging the gap between latest machine learning research and industry applications in microscopy

Abstract
In recent years, neural network based image enhancement approaches have made significant progress for advancing the generation and analysis of image data in microscopy. E.g. machine learning based denoising methods, such as Noise2Void (N2V), are applicable to a large variety of noisy datasets, even without dedicated training data being available. Arguably, this facilitates the fast and widespread adoption of machine learning based methods throughout the microscopy community. Unfortunately, the blind-spot training underlying N2V can lead to rather visible checkerboard artifacts compromising the image quality. In this presentation, we will show approaches how to overcome these obstacles and provide fast and reliable solutions to our ZEISS microscopy customers.

Bio
Dr. Tanja Teuber is heading the department for algorithm development in the Corporate Research & Technology unit of Carl Zeiss. In her role, she is responsible for multiple teams and 35 research scientists driving research and development projects in the fields of machine learning, computer vision, computational imaging and optics simulation.
Before joining Carl Zeiss in 2013, she studied Mathematics and Computer Science at the University of Mannheim and Swansea University. Moreover, she completed a PhD in Mathematical Image Processing at the University of Kaiserslautern with distinction in 2012.